system

Abstract

We provide the system. [Solution] A means for inputting user voice and converting said voice into a digital format, A speech recognition means that converts the audio converted to the aforementioned digital format into text, A translation means for translating the aforementioned text into a predetermined foreign language, A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data, A system including a playback means for playing the aforementioned audio data to a user.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Many people have problems communicating in foreign languages. However, using professional simultaneous interpreters is costly and difficult to use constantly. Also, although there are automatic translation applications, they have limitations in real-time translation performance and cannot smoothly progress conversations. To solve such problems, there is a demand for a system that can perform translation and speech synthesis easily, at low cost, in real time, and enable smooth communication.

Means for Solving the Problems

[0005] To solve the above problems, the present invention provides the following means: a system including means for inputting user voice and converting said voice into a digital format; voice recognition means for converting the converted voice into text; translation means for translating said text into a predetermined foreign language; voice synthesis means for converting said text into audio data; and playback means for playing said audio data to the user. The voice recognition means uses a cloud-based voice recognition engine to perform the entire process quickly and accurately. Furthermore, by using the speakers of a smartphone or tablet as the playback means, a system that can be easily used by users anywhere is realized.

[0006] A "user" is a person who uses the system to perform voice input.

[0007] "Voice input" refers to the process of converting the words a user speaks to a system into a digital format.

[0008] "Digital format" refers to the conversion of analog audio data into digital data that can be processed by a computer.

[0009] "Speech recognition means" refers to technology that converts input digital audio into text data.

[0010] "Translation means" refers to the process of converting data that has been transcribed into text by speech recognition means into another language.

[0011] "Speech synthesis means" refers to a technology that converts text data, which has been transformed by translation means, into speech data.

[0012] "Playback means" refers to a device or function that plays back audio data generated by speech synthesis means in a way that the user can hear.

[0013] A "cloud-based speech recognition engine" is a speech recognition service that can be used over the internet and has the function of receiving speech data from multiple devices and converting it into text data.

[0014] A "speaker" is an output device that reproduces audio data as physical sound. [Brief explanation of the drawing]

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

Mode for Carrying Out the Invention

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

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

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

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

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

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

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

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

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

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

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

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

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

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

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

[0036] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. The system's program processing will be described in natural language to illustrate embodiments of the invention. Specific usage examples will also be provided.

[0037] System Overview

[0038] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system consists of the following main elements:

[0039] 1. Voice input method

[0040] 2. Speech recognition means

[0041] 3. Translation methods

[0042] 4. Speech synthesis means

[0043] 5. Regeneration means

[0044] System processing flow

[0045] 1. Voice input

[0046] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[0047] 2. Speech Recognition

[0048] Terminal: Converts voice input to a digital format and saves it locally.

[0049] Terminal: Sends stored digital audio data to a cloud-based speech recognition engine.

[0050] Server: The speech recognition engine converts the digital speech data into text data "Hello, what is your name?".

[0051] 3. Text Translation

[0052] Server: Sends the speech-recognized text to the translation engine.

[0053] Server: The translation engine translates the text into English as "Hello, what is your name?".

[0054] 4. Speech synthesis

[0055] Server: Sends the translated English text "Hello, what is your name?" to the speech synthesis engine.

[0056] Server: The speech synthesis engine converts English text into speech data.

[0057] Server: Sends the generated audio data to the terminal.

[0058] 5. Audio Playback

[0059] Terminal: Receives audio data and plays it back to the user.

[0060] Specific example

[0061] For example, the specific process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0062] 1. User: "Hello, what's your name?" they say to their smartphone.

[0063] 2. Device: Record audio, convert it to a digital format, and save it.

[0064] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[0065] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[0066] 5. Server: The translation engine translates the text into English as "Hello, what is your name?".

[0067] 6. Server: The speech synthesis engine converts English text into speech data.

[0068] 7. Server: Sends audio data to the user's terminal.

[0069] 8. Terminal: Play the audio data "Hello, what is your name?" to the user.

[0070] 9. The other person responds, "My name is John."

[0071] 10. The other party's device: Record the audio, convert it to a digital format, and save it.

[0072] 11. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[0073] 12. Server: The speech recognition engine converts "My name is John." into text.

[0074] 13. Server: The translation engine translates the text into Japanese as "My name is John."

[0075] 14. Server: The speech synthesis engine converts Japanese text into speech data.

[0076] 15. Server: Sends audio data to the user's terminal.

[0077] 16. Terminal: Play the audio data "My name is John." to the user.

[0078] The above process enables smooth communication between users and people who speak a foreign language.

[0079] The following describes the processing flow.

[0080] Step 1:

[0081] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[0082] Step 2:

[0083] Terminal: The voice input application records the user's speech and converts it into a digital format.

[0084] Step 3:

[0085] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[0086] Step 4:

[0087] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[0088] Step 5:

[0089] Server: Sends the speech-recognized text data to the translation engine.

[0090] Step 6:

[0091] Server: The translation engine translates the text data into English as "Hello, what is your name?".

[0092] Step 7:

[0093] Server: Sends the translated English text data to the speech synthesis engine.

[0094] Step 8:

[0095] Server: The speech synthesis engine converts English text data into speech data.

[0096] Step 9:

[0097] Server: Sends the generated audio data to the terminal.

[0098] Step 10:

[0099] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio "Hello, what is your name?".

[0100] Step 11:

[0101] The other party (device): The other party responds with "My name is John."

[0102] Step 12:

[0103] Recipient's device: Records the recipient's speech and converts it to a digital format.

[0104] Step 13:

[0105] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[0106] Step 14:

[0107] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[0108] Step 15:

[0109] Server: Sends the speech-recognized text data to the translation engine.

[0110] Step 16:

[0111] Server: The translation engine translates the text data into Japanese as "My name is John."

[0112] Step 17:

[0113] Server: Sends the translated Japanese text data to the speech synthesis engine.

[0114] Step 18:

[0115] Server: The speech synthesis engine converts Japanese text data into speech data.

[0116] Step 19:

[0117] Server: Sends the generated audio data to the user's device.

[0118] Step 20:

[0119] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio saying, "My name is John."

[0120] This series of steps enables real-time voice translation between the user and the person speaking the foreign language, facilitating smooth communication.

[0121] (Example 1)

[0122] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0123] In modern society, smooth communication between people who speak different languages ​​is a challenging task. In particular, there is a lack of technology for real-time translation and seamless multilingual conversation. Therefore, there is a need to develop systems that allow users who speak different languages ​​to communicate easily.

[0124] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0125] In this invention, the server includes means for inputting the user's voice and converting the voice into a digital format; speech recognition means for converting the digitally converted voice into text; translation means for translating the text into a predetermined foreign language; speech synthesis means for converting the translated text into audio data; playback means for playing the audio data to the user; means for transmitting the text data obtained from the speech recognition means to a translation engine; means for transmitting the translated text obtained from the translation engine to a speech synthesis engine; and means for transmitting the audio data obtained from the speech synthesis engine to the user's terminal. This enables users who speak different languages ​​to communicate smoothly in real time.

[0126] A "user" refers to a person who uses a system to communicate.

[0127] "Means of voice input" refers to devices and technologies for converting a user's voice into a digital format.

[0128] "Digital format" refers to a state in which audio or data has been converted into a format that can be processed by electronic devices.

[0129] "Speech recognition means" refers to a function or technology that converts speech, which has been converted into a digital format, into text.

[0130] "Translation means" refers to a function or technology for translating text data into a specified foreign language.

[0131] "Speech synthesis means" refers to a function or technology for converting text data translated into a foreign language into speech data.

[0132] "Playback means" refers to devices and technologies for providing audio data to users as auditory information.

[0133] A "cloud-based speech recognition engine" refers to a speech recognition service provided via the internet.

[0134] "User's personal information terminal" refers to portable electronic devices such as smartphones and tablets.

[0135] "Audio output device" refers to a device that includes speakers or headphones for playing back audio data.

[0136] "Text data" refers to standard string information converted by speech recognition.

[0137] "Translation" refers to the text data after it has been translated.

[0138] "Audio data" refers to the format of digital audio generated by speech synthesis technology.

[0139] This invention is a system aimed at enabling users who speak different languages ​​to communicate smoothly in real time. The system includes a series of processes that convert audio into a digital format, convert it to text, translate it, convert it back into audio data, and finally play it back to the user.

[0140] System Configuration

[0141] This system consists of the following main elements:

[0142] 1. Voice input method

[0143] 2. Speech recognition means

[0144] 3. Translation methods

[0145] 4. Speech synthesis means

[0146] 5. Regeneration means

[0147] Hardware and software to use

[0148] For voice input, the user's mobile device, such as a smartphone or tablet, is used. These devices have built-in microphones that record the user's voice.

[0149] The speech recognition method sends recorded audio to a cloud-based speech recognition engine, which then converts the audio to text. Specifically, it uses services such as Google's Cloud Speech-to-Text API.

[0150] The translation method translates the text obtained by the speech recognition method into a specified foreign language. This uses a translation engine such as the Google Cloud Translation API.

[0151] Speech synthesis methods convert translated text into speech data. A specific example used is the Microsoft® Azure® Text-to-Speech API.

[0152] The playback method involves playing the generated audio data to the user. The speaker or earphones built into the user's smartphone or tablet are used as the playback device.

[0153] Specific example

[0154] For example, the process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0155] 1. The user says to their smartphone, "Hello, what's your name?"

[0156] 2. The device records the audio, converts it to a digital format, and saves it.

[0157] 3. The device sends the digital voice data to a cloud-based speech recognition engine (Google Cloud Speech-to-Text API).

[0158] 4. The server converts the audio data into text data.

[0159] 5. The server sends the text data to the translation engine (Google Cloud Translation API) for translation into English.

[0160] 6. The server sends the English text data to the text-to-speech engine (Microsoft Azure Text-to-Speech API) and converts it into speech data.

[0161] 7. The server sends the generated English audio data to the user's device.

[0162] 8. The device plays audio data, and an English voice says, "Hello, what is your name?"

[0163] Example of a prompt

[0164] The following are specific examples of prompt statements used in this system.

[0165] Prompt: Translate the Japanese phrase "Hello, what is your name?" into English, and then generate the English audio.

[0166] Using this prompt, the system goes through the processes of speech recognition, text translation, and speech synthesis to generate and provide results to the user.

[0167] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0168] Step 1:

[0169] Voice input

[0170] The user speaks into their smartphone or tablet.

[0171] Input: User's voice (e.g., "Hello, what's your name?")

[0172] Specific action: The user speaks into the microphone.

[0173] Output: Raw audio data

[0174] Step 2:

[0175] Audio recording and digital conversion

[0176] The device records the user's voice and converts it to a digital format.

[0177] Input: Raw audio data

[0178] Specific operation: The smartphone's audio engine (e.g., AudioRecord API) converts the raw audio into PCM digital data.

[0179] Output: Digital audio data in PCM format

[0180] Step 3:

[0181] Sending audio data

[0182] The device sends the converted digital audio data to a cloud-based speech recognition engine.

[0183] Input: PCM format digital audio data

[0184] Specific operation: Send digital audio data as an HTTP request to the Google Cloud Speech-to-Text API.

[0185] Output: Audio data sent to the server

[0186] Step 4:

[0187] Speech recognition

[0188] The server uses a speech recognition engine to convert digital audio data into text data.

[0189] Input: PCM format digital audio data

[0190] Specific operation: The Google Cloud Speech-to-Text API analyzes the audio data and converts it into text, "Hello, what is your name?".

[0191] Output: Text data "Hello, what is your name?"

[0192] Step 5:

[0193] Text translation

[0194] The server sends the text data acquired through speech recognition to the translation engine.

[0195] Input: Text data "Hello, what is your name?"

[0196] Specific operation: The server sends a translation request to the Google Cloud Translation API. The request includes the source text and the target language.

[0197] Output: English translation text data "Hello, what is your name?"

[0198] Step 6:

[0199] Preparation for speech synthesis

[0200] The server sends the translated English text to the speech synthesis engine.

[0201] Input: English text data "Hello, what is your name?"

[0202] Specific operation: The server sends a text-to-speech request to the Microsoft Azure Text-to-Speech API. The request includes the translated text and voice configuration information.

[0203] Output: Voice data request

[0204] Step 7:

[0205] Speech synthesis

[0206] The server uses a speech synthesis engine to convert text data into speech data.

[0207] Input: English text data "Hello, what is your name?"

[0208] Specific operation: The Microsoft Azure Text-to-Speech API parses English text and converts it into speech data.

[0209] Output: English audio data

[0210] Step 8:

[0211] Sending audio data

[0212] The server sends the generated English audio data to the user's device.

[0213] Input: English audio data

[0214] Specific operation: The server sends the generated audio data to the terminal as an HTTP response.

[0215] Output: Audio data sent to the user's terminal

[0216] Step 9:

[0217] Playback of audio data

[0218] The device plays the received audio data.

[0219] Input: English audio data

[0220] Specific action: The device's media player (e.g., MediaPlayer API) plays audio data, and the voice "Hello, what is your name?" is played through the speaker.

[0221] Output: Audio played to the user

[0222] (Application Example 1)

[0223] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0224] In brick-and-mortar retail environments, where smooth communication between customers and employees with different language backgrounds is essential, language barriers can prevent customers from obtaining necessary information, leading to a decline in service quality. Furthermore, the difficulty for employees to be fluent in multiple languages ​​reduces operational efficiency in physical stores. A system is needed to address these challenges and support communication in brick-and-mortar retail settings.

[0225] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0226] In this invention, the server includes means for inputting a user's voice and converting the voice into a digital format; voice recognition means for converting the voice converted into digital format into text; translation means for translating the text into a predetermined foreign language; voice synthesis means for converting the text translated into the foreign language into audio data; playback means for playing the audio data back to the user; means for supporting communication between customers and employees in a store; and means for customers or employees to speak into the device. This enables real-time voice translation and smooth communication between customers and employees who speak different languages.

[0227] "User" refers to the customers and employees who use this system.

[0228] "Means for inputting audio and converting said audio into a digital format" refers to a device or software that has the function of converting analog audio data into digital data.

[0229] "Speech recognition means" refers to a device or software that has the function of converting speech converted into digital format into text format.

[0230] "Translation means" refers to a device or software that has the function of translating text, which has been converted into text format by speech recognition means, into a predetermined foreign language.

[0231] "Speech synthesis means" refers to a device or software that has the function of converting translated text into speech data.

[0232] "Playback means" refers to a device or software that has the function of playing back audio data generated by a speech synthesis means to the user.

[0233] "Means for supporting communication between customers and employees within a store" refers to devices or software that provide the necessary functions for smooth communication between customers and employees who speak different languages ​​in a physical store.

[0234] "Device" refers to a personal information terminal (such as a smartphone or tablet) used for inputting and playing audio.

[0235] This invention is a multilingual real-time interpretation assistant system aimed at facilitating smooth communication between customers and employees who speak different languages. This invention is primarily implemented using mobile devices such as smartphones and tablets, and utilizes a cloud-based speech recognition engine and translation engine.

[0236] System Configuration

[0237] This system includes the following key hardware and software components:

[0238] 1. Mobile information terminal

[0239] It has a microphone and processor for voice input, recording, and conversion to digital format.

[0240] 2. Cloud-based speech recognition engine

[0241] Use the Google Cloud Speech-to-Text API to convert digital audio data into text data.

[0242] 3. Translation engine

[0243] The retrieved text data is translated into a specified foreign language using the Google Cloud Translate API.

[0244] 4. Speech synthesis engine

[0245] The Google Cloud Text-to-Speech API is used to convert translated text data into audio data.

[0246] 5. Regeneration means

[0247] Audio data is played using the speaker installed in the mobile device.

[0248] Processing flow details

[0249] The server includes means for inputting user voice and converting said voice into a digital format, voice recognition means, translation means, speech synthesis means, and playback means. By coordinating these means, real-time communication between customers and employees in a physical store is realized.

[0250] 1. Voice input and conversion

[0251] Personal digital assistant (PDA): The user (customer or employee) speaks into the device. The device records the audio and converts it to a digital format.

[0252] 2. Speech Recognition

[0253] Mobile device: Sends digital audio data to the Google Cloud Speech-to-Text API. The Google Cloud Speech-to-Text API converts the audio data into text data.

[0254] 3. Text Translation

[0255] Server: Sends text data to the Google Cloud Translate API for translation into the specified foreign language.

[0256] 4. Speech synthesis

[0257] Server: Sends the translated text data to the Google Cloud Text-to-Speech API and converts it into speech data.

[0258] 5. Audio Playback

[0259] Personal digital assistant (PDCA): Receives generated audio data and plays it through the speaker.

[0260] Specific example

[0261] For example, consider a case where a Japanese-speaking customer asks an English-speaking employee for the location of a product.

[0262] 1. Customer: "Where can I find this product?" they say to their smartphone.

[0263] 2. Personal digital assistant (PDCA): Records audio, converts it to a digital format, and sends it to the cloud.

[0264] 3. Google Cloud Speech-to-Text API: Converts audio data into text data such as "Where can I find this product?".

[0265] 4. Google Cloud Translate API: Translate the text data into "Where is this product located?".

[0266] 5. Google Cloud Text-to-Speech API: Converts translated text data into speech data.

[0267] 6. Mobile device: Play the English audio "Where is this product located?".

[0268] Example of a prompt

[0269] "Where can I find this product?"

[0270] In this way, customers and employees can communicate smoothly, overcoming language barriers.

[0271] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0272] Step 1:

[0273] Voice input and conversion to digital format

[0274] User: The user (customer or employee) speaks towards the mobile information terminal.

[0275] Input: Analog voice data.

[0276] Terminal: The microphone of the mobile information terminal records the voice and converts the analog voice data into digital format.

[0277] Output: Digital voice data.

[0278] Step 2:

[0279] Voice recognition of digital voice data

[0280] Terminal: Transmits the digital voice data to the cloud.

[0281] Input: Digital voice data.

[0282] Server (Google Cloud Speech-to-Text API): Receives the digital voice data and performs voice recognition. Specifically, it analyzes the voice pattern and converts it into string data.

[0283] Output: Text data (e.g., "Where is this product?").

[0284] Step 3:

[0285] Translation of text data

[0286] Server: Translates the text data into the specified foreign language using a translation engine (Google Cloud Translate API).

[0287] Input: Japanese text data (e.g., "Where can I find this product?").

[0288] Server (Google Cloud Translate API): Receives text data and performs translation processing. Based on the model, it translates Japanese into the target language, such as English.

[0289] Output: Translated text data (e.g., "Where is this product located?").

[0290] Step 4:

[0291] Converting translated text to audio data

[0292] Server: Sends the translated text data to the text-to-speech engine (Google Cloud Text-to-Speech API) and converts it into speech data.

[0293] Input: Translated text data (e.g., "Where is this product located?").

[0294] Server (Google Cloud Text-to-Speech API): Converts text data into speech data. Specifically, it analyzes the text, breaks it down into phonemes, and then synthesizes them together.

[0295] Output: Audio data (e.g., "Where is this product located?").

[0296] Step 5:

[0297] Playback of audio data

[0298] Terminal: Receives audio data and plays it back to the user.

[0299] Input: Voice data received from the server (e.g., "Where is this product located?").

[0300] Terminal (Speaker): Plays the voice data.

[0301] Output: Response message by voice (e.g., "Where is this product located?").

[0302] Through this processing step, the user can interact with a person who speaks a different language in real time.

[0303] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0304] This invention relates to a system that facilitates communication in a foreign language and translates voices in real time. Also provided is a system that combines an emotion engine that recognizes the user's emotion and appropriately adjusts the translation expression according to the emotion. Regarding the form for implementing this invention, the processing of the system program is described in natural language. Also, it is described with specific usage examples.

[0305] Overview of the System

[0306] The system of this invention is designed for the user to communicate smoothly with a person who speaks a foreign language. This system includes the following main elements.

[0307] 1. Voice input means

[0308] 2. Voice recognition means

[0309] 3. Translation means

[0310] 4. Emotion engine

[0311] 5. Speech synthesis means

[0312] 6. Reproduction means

[0313] System processing flow

[0314] 1. Voice input

[0315] User (terminal): The user speaks towards a smartphone or tablet. For example, the user makes a statement like "Hello, what's your name?".

[0316] 2. Speech recognition

[0317] Terminal: Convert the voice input into digital format and save it locally.

[0318] Terminal: Transmit the saved digital voice data to a cloud-based speech recognition API.

[0319] Server: The speech recognition engine converts the digital voice data into text data "Hello, what's your name?".

[0320] 3. Emotion recognition

[0321] Server: The emotion engine analyzes the emotion from the user's voice in real time and generates emotion data. For example, it recognizes that the statement "Hello, what's your name?" is in a happy tone.

[0322] 4. Text translation

[0323] Server: Transmit the speech-recognized text data and emotion data to the translation engine.

[0324] Server: The translation engine translates the text data "Hello, what's your name?" into "Hello, what is your name?" in English and adjusts the expression to fit the user's emotion.

[0325] 5. Speech synthesis

[0326] Server: Sends the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[0327] Server: The speech synthesis engine converts English text data into speech data.

[0328] Server: Sends the generated audio data to the terminal.

[0329] 6. Audio Playback

[0330] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play a cheerful voice saying, "Hello, what is your name?"

[0331] Specific example

[0332] For example, the processing flow when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0333] 1. User: "Hello, what's your name?" they say to their smartphone.

[0334] 2. Device: Record audio, convert it to a digital format, and save it.

[0335] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[0336] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[0337] 5. Server: The emotion engine recognizes emotions from the user's voice and generates "happy" emotion data.

[0338] 6. Server: The translation engine translates the text into English as "Hello, what is your name?", maintaining the cheerful tone.

[0339] 7. Server: The speech synthesis engine converts English text into speech data.

[0340] 8. Server: Sends audio data to the user's terminal.

[0341] 9. Device: Play the voice data "Hello, what is your name?" in a cheerful tone.

[0342] 10. The other person responds, "My name is John."

[0343] 11. The other party's device: Record the audio, convert it to a digital format, and save it.

[0344] 12. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[0345] 13. Server: The speech recognition engine converts "My name is John." into text.

[0346] 14. Server: The emotion engine recognizes emotions from the other party's voice and generates "calm" emotion data.

[0347] 15. Server: The translation engine translates the text into Japanese as "My name is John," maintaining a calm tone.

[0348] 16. Server: The speech synthesis engine converts Japanese text into speech data.

[0349] 17. Server: Sends audio data to the user's terminal.

[0350] 18. Device: Play the audio data "My name is John." in a calm tone.

[0351] In this way, smooth and considerate communication between the user and the person speaking the foreign language is achieved.

[0352] The following describes the processing flow.

[0353] Step 1:

[0354] User (device): The user speaks to their smartphone or tablet and says, "Hello, what's your name?"

[0355] Step 2:

[0356] Terminal: The voice input application records the user's speech and converts it into a digital format.

[0357] Step 3:

[0358] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[0359] Step 4:

[0360] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[0361] Step 5:

[0362] Server: Sends text data to the emotion engine to analyze the user's emotions.

[0363] Step 6:

[0364] Server: The emotion engine analyzes the tone and pitch of the voice, etc., and recognizes that the user has a "happy" emotion. Also, it generates emotion data.

[0365] Step 7:

[0366] Server: Transmits the voice-recognized text data and emotion data to the translation engine.

[0367] Step 8:

[0368] Server: The translation engine translates the text data "こんにちは、お名前は何ですか?" into English as "Hello, what is your name?" and adjusts the expression to fit the user's "happy" emotion.

[0369] Step 9:

[0370] Server: Transmits the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[0371] Step 10:

[0372] Server: The speech synthesis engine converts the English text data into speech data.

[0373] Step 11:

[0374] Server: Transmits the generated speech data to the terminal.

[0375] Step 12:

[0376] Terminal: Receives the speech data and plays it as speech to the user. For example, the speech "Hello, what is your name?" is played in a happy tone.

[0377] Step 13:

[0378] The other party (device): The other party responds with "My name is John."

[0379] Step 14:

[0380] Recipient's device: Records the recipient's speech and converts it to a digital format.

[0381] Step 15:

[0382] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[0383] Step 16:

[0384] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[0385] Step 17:

[0386] Server: Sends text data to the emotion engine to analyze the other party's emotions.

[0387] Step 18:

[0388] Server: The emotion engine analyzes the tone and pitch of the voice and recognizes that the other party has a "calm" emotion. It also generates emotion data.

[0389] Step 19:

[0390] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[0391] Step 20:

[0392] Server: The translation engine translates the text data "My name is John." into Japanese as "私の名前はジョンです。" and adjusts the expression to fit the "calm" emotion of the other party.

[0393] Step 21:

[0394] Server: The translated Japanese text data "私の名前はジョンです。" is sent to the text-to-speech engine.

[0395] Step 22:

[0396] Server: The text-to-speech engine converts the Japanese text data into audio data.

[0397] Step 23:

[0398] Server: The generated audio data is sent to the user's terminal.

[0399] Step 24:

[0400] Terminal: Receives the audio data and plays it as audio for the user. For example, the audio is played in a calm tone as "私の名前はジョンです。".

[0401] In this way, smooth communication that takes into account emotions is realized between the user and the foreign language speaker.

[0402] (Example 2)

[0403] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0404] In recent years, international communication has increased, and smooth multilingual conversation is required. However, conventional translation systems simply translate languages ​​without adjusting the translation to take into account the speaker's emotions, resulting in a decline in the quality of communication. Furthermore, in real-time speech translation, the processing speed from speech recognition to translation and playback has been a problem. Therefore, there is a need for a system that accurately reflects the speaker's emotions while enabling smooth, real-time communication between multiple languages.

[0405] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0406] In this invention, the server includes means for recognizing emotions from the user's voice in real time and generating emotion data, means for adjusting translated expressions based on translated text and emotion data, and means for converting text translated into a foreign language into speech data. This enables real-time multilingual voice communication using translated expressions that reflect the speaker's emotions.

[0407] A "user" refers to a person who uses the system to perform voice input and voice translation.

[0408] "Digital format" refers to a format in which analog audio is converted into digital data that can be processed by a computer.

[0409] "Speech recognition means" refers to means that have the function of converting digital speech data into text data.

[0410] "Translation means" refers to a means that has the function of converting text from one language to another language.

[0411] "Emotion recognition means" refers to means that have the function of analyzing the speaker's emotions in real time from audio data and generating emotion data.

[0412] "Means for adjusting translated expressions" refers to means that have the function of adjusting translated expressions to reflect appropriate sentiment based on the translated text and sentiment data.

[0413] "Speech synthesis means" refers to means that have the function of converting text data into speech data.

[0414] "Playback means" refers to means that have the function of outputting the generated audio data as sound.

[0415] A "cloud-based speech recognition engine" refers to an engine that uses servers and software accessible via a network to convert speech data into text data.

[0416] "Communication equipment" refers to electronic devices with communication capabilities, such as smartphones and tablets.

[0417] "Audio equipment" refers to devices such as speakers and earphones used to play audio data.

[0418] This invention is a speech translation system that enables users to communicate smoothly with people who speak a foreign language. The system consists of the following processes: speech input, speech recognition, emotion recognition, text translation, speech synthesis, and speech playback. Each process utilizes specific hardware and software.

[0419] System Configuration

[0420] 1. Voice input

[0421] User: Speaks into a smartphone or tablet. For example, says, "Hello, what's your name?"

[0422] Terminal: Uses a microphone as an interface to convert the user's voice into a digital format.

[0423] 2. Speech Recognition

[0424] Terminal: Sends digital voice data to a cloud-based speech recognition engine. The API used is the speech recognition engine.

[0425] Server: Converts the digital audio data received by the speech recognition engine into text data.

[0426] 3. Emotion recognition

[0427] Server: The emotion recognition engine analyzes the user's voice in real time and generates emotion data. The emotion recognition engine used is an emotion analysis engine.

[0428] 4. Text Translation

[0429] Server: Sends the speech-recognized text data and generated sentiment data to the translation engine. The translation engine used is a text translation engine.

[0430] Server: The translation engine converts the text into the target language and adjusts the translation based on sentiment data.

[0431] 5. Speech synthesis

[0432] Server: Sends the translated text data to the speech synthesis engine. The speech synthesis engine used is speech synthesis software.

[0433] Server: The speech synthesis engine converts text data into speech data.

[0434] 6. Audio Playback

[0435] Terminal: Receives audio data and plays it back using the terminal's sound device. For example, it might play back "Hello, what is your name?" in a cheerful tone.

[0436] Specific example

[0437] As a concrete example, we will simulate a conversation between a Japanese-speaking user and an English-speaking partner. The following is the processing flow.

[0438] 1. User: "Hello, what's your name?" they say to their smartphone.

[0439] 2. Device: Records audio and converts it into digital data.

[0440] 3. Server: The speech recognition engine converts "Hello, what is your name?" into text data.

[0441] 4. Server: The emotion recognition engine generates text data and emotion data, and detects the emotion of "happiness".

[0442] 5. Server: The translation engine translates the text data into English as "Hello, what is your name?", maintaining a cheerful tone.

[0443] 6. Server: The speech synthesis engine converts the translated text into speech data.

[0444] 7. Terminal: Receives audio data and plays "Hello, what is your name?" in a cheerful tone.

[0445] Example of a prompt

[0446] Use the following prompts to input data into the generative AI model.

[0447] "Please translate the Japanese phrase 'Hello, what is your name?' into English and convert it into speech, while maintaining the emotion in the speech."

[0448] "Translate the English sentence 'My name is John.' into Japanese and then speak it in a calm tone."

[0449] In this way, real-time voice translation that takes emotions into consideration is achieved between the user and the person speaking the foreign language.

[0450] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0451] Program processing flow

[0452] Step 1: Voice Input

[0453] User: Speak into the microphone on your smartphone or tablet and say, "Hello, what's your name?"

[0454] Input: Analog audio

[0455] Output: Digital audio data

[0456] Terminal: Captures the user's voice with a microphone, converts the analog audio to a digital format using an audio digitization module, and saves it to a buffer.

[0457] Step 2: Speech Recognition

[0458] Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[0459] Input: Digital audio data

[0460] Output: Text data ("Hello, what is your name?")

[0461] Server: The speech recognition engine analyzes the digital speech data it receives and converts it into text data, taking into account phonemes and context.

[0462] Step 3: Emotion Recognition

[0463] Server: Sends the speech-recognized text data and digital audio data to the emotion recognition engine.

[0464] Input: Text data, digital audio data

[0465] Output: Emotional data (e.g., "Happy")

[0466] Server: The emotion recognition engine analyzes the tone, pitch, and speed of the voice, evaluates the user's emotional state in real time, and generates emotion data.

[0467] Step 4: Text Translation

[0468] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[0469] Input: Text data, sentiment data

[0470] Output: Translated text data (e.g., "Hello, what is your name?")

[0471] Server: The translation engine converts text data into the target foreign language and adjusts the translation based on sentiment data.

[0472] Step 5: Speech Synthesis

[0473] Server: Sends the translated text data to the speech synthesis engine.

[0474] Input: Translated text data

[0475] Output: Audio data

[0476] Server: The speech synthesis engine analyzes the translated text data and generates speech data that reflects emotional tone.

[0477] Step 6: Audio Playback

[0478] Device: Receives audio data and plays it through the device's speaker.

[0479] Input: Audio data

[0480] Output: Played audio (e.g., "Hello, what is your name?")

[0481] Terminal: Uses an audio device to play the generated audio data to the user.

[0482] (Application Example 2)

[0483] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0484] In physical stores, communication between customers who speak different languages ​​and employees often breaks down. This can lead to decreased customer satisfaction and a decline in the quality of service. Furthermore, it is difficult for employees to immediately understand and respond to multiple languages, especially when it comes to conveying nuances, including emotions. In this situation, it is necessary to improve customer service in physical stores by streamlining foreign language support and providing accurate translations that take emotions into consideration.

[0485] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0486] This invention includes a server that includes means for inputting a user's voice and converting the voice into a digital format, speech recognition means for converting the digitally converted voice into text, translation means for translating the text into a predetermined foreign language, an emotion engine that recognizes the emotion of the translated text and adjusts the translated expression appropriately according to the emotion, speech synthesis means for converting the translated text into audio data, playback means for playing the audio data back to the user, and a device that implements an application that supports real-time multilingual communication between the user and customers who speak a foreign language. This makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the emotions of customers.

[0487] "Means of inputting audio and converting it to a digital format" refers to devices or processes that receive audio signals and record them as digital data.

[0488] "Speech recognition means" refers to a technology or device that analyzes speech converted into digital data and converts it into corresponding text data.

[0489] "Translation methods" refer to software or algorithms used to convert text data from one language into another language.

[0490] An "emotion engine" is a technology or algorithm that recognizes a speaker's emotions from audio or text data and analyzes that emotional information.

[0491] "Speech synthesis means" refers to a technology or device that analyzes text data and generates speech data corresponding to that text.

[0492] "Reproduction means" refers to a device or technology used to actually pronounce the generated audio data as sound.

[0493] "Device" is a general term for any equipment on which an application is installed, such as communication terminals, smartphones, and smart glasses.

[0494] A "cloud-based speech recognition engine" refers to a service or technology for performing speech recognition in a cloud computing environment.

[0495] A "communication terminal" refers to an electronic device such as a smartphone or tablet that has functions such as voice input and playback, and data communication.

[0496] An "application that supports multilingual communication" is software designed to enable users who speak different languages ​​to communicate with each other in real time.

[0497] This invention is a system that facilitates multilingual support in physical stores and provides appropriate translations that take emotions into consideration. This system uses a device to support real-time communication between users and foreign language-speaking customers. The main elements of the system are means for inputting voice and converting it to a digital format, voice recognition means, translation means, emotion engine, voice synthesis means, playback means, and a device for implementing the application.

[0498] The following hardware and software will be used to implement this system:

[0499] Hardware: Communication devices (e.g., smartphones, tablets, smart glasses)

[0500] software:

[0501] Speech recognition: Google Cloud Speech-to-Text API

[0502] Translation: Google Cloud Translation API

[0503] Emotion recognition: IBM Watson(R) Tone Analyzer API

[0504] Text-to-speech: Google Cloud Text-to-Speech API

[0505] Real-time communication: Firebase Realtime Database

[0506] Description of the system's program processing

[0507] First, the user speaks into a communication device. The communication device converts this audio into a digital format and uploads it to Firebase. The server sends the audio data stored in Firebase to the Google Cloud Speech-to-Text API, where it converts the audio into text data.

[0508] Next, the server sends this text data to the IBM Watson Tone Analyzer API for sentiment analysis. The text, including sentiment data, is then sent to the Google Cloud Translation API for translation into the specified foreign language. Based on the sentiment data, the translated sentence is adjusted to an appropriate tone.

[0509] The server then sends the translated text data to the Google Cloud Text-to-Speech API to generate audio data. This audio data is then stored again in Firebase and sent to the user's communication device. The communication device plays this audio data, allowing the user to provide the translated message to their customers.

[0510] Specific example

[0511] Let's consider a situation where an employee working in a souvenir shop in a tourist area needs to communicate smoothly with foreign customers. For example,

[0512] User (employee): "Welcome, how can I help you?" they say to their smartphone.

[0513] System: The server recognizes the speech, analyzes the emotions, translates it into English, and plays back "Welcome! How can I help you?" in a friendly tone.

[0514] The other party (customer) asks, "Can you tell me where I can find local souvenirs?"

[0515] User: Provide a properly translated response through the system again to facilitate smooth communication.

[0516] Example of a prompt

[0517] "Based on the following information, please generate code that performs translations tailored to the customer's emotions."

[0518] Customer comment: "Hello, can I purchase this?"

[0519] Emotion: "Friendly"

[0520] Translated statement: "Hello, can I purchase this?"

[0521] Code for speech synthesis in a friendly tone

[0522] This invention makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the customer's feelings.

[0523] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0524] Step 1:

[0525] Input: The user speaks into the communication terminal.

[0526] Operation: The user speaks to the communication terminal saying, "Welcome, how can I help you?"

[0527] Data processing: The communication terminal converts this audio into a digital format.

[0528] Output: Digital audio data is generated.

[0529] Step 2:

[0530] Input: Digital audio data.

[0531] Operation: The communication device uploads this digital audio data to Firebase.

[0532] Data processing: Digital audio data is transferred to the cloud.

[0533] Output: Digital audio data is stored on Firebase.

[0534] Step 3:

[0535] Input: Digital audio data stored in Firebase.

[0536] Operation: The server retrieves audio data from Firebase and sends it to the Google Cloud Speech-to-Text API.

[0537] Data processing: The speech recognition engine analyzes the digital audio data and converts it into text data.

[0538] Output: The text data "Welcome, how can I help you?" is generated.

[0539] Step 4:

[0540] Input: Text data.

[0541] Operation: The server sends text data to the IBM Watson Tone Analyzer API.

[0542] Data processing: The emotion engine analyzes the text data and generates emotion data (e.g., "friendly tone").

[0543] Output: Sentiment data is generated and attached to the text data.

[0544] Step 5:

[0545] Input: Text data and sentiment data.

[0546] Operation: The server sends this data to the Google Cloud Translation API.

[0547] Data processing: The translation engine translates text data into the specified foreign language and adjusts it based on sentiment data.

[0548] Output: The translated text data "Welcome! How can I help you?" is generated.

[0549] Step 6:

[0550] Input: Translated text data.

[0551] Operation: The server sends this text data to the Google Cloud Text-to-Speech API.

[0552] Data processing: The speech synthesis engine converts the translated text data into speech data.

[0553] Output: The audio data "Welcome! How can I help you?" is generated.

[0554] Step 7:

[0555] Input: Translated audio data.

[0556] Operation: The server uploads the generated audio data to Firebase and sends it to the user's communication device.

[0557] Data processing: Audio data is transferred from the server to the communication terminal.

[0558] Output: The audio data is saved to the communication terminal.

[0559] Step 8:

[0560] Input: Translated audio data.

[0561] Operation: The communication terminal plays the translated audio data through its speaker.

[0562] Data processing: Audio data is output as audio.

[0563] Output: The translated audio "Welcome! How can I help you?" is played to the customer.

[0564] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0565] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0566] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0567] [Second Embodiment]

[0568] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0569] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0570] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0571] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0572] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0573] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0574] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0575] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0576] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0577] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0578] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0579] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0580] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. The system's program processing will be described in natural language to illustrate embodiments of the invention. Specific usage examples will also be provided.

[0581] System Overview

[0582] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system consists of the following main elements:

[0583] 1. Voice input method

[0584] 2. Speech recognition means

[0585] 3. Translation methods

[0586] 4. Speech synthesis means

[0587] 5. Regeneration means

[0588] System processing flow

[0589] 1. Voice input

[0590] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[0591] 2. Speech Recognition

[0592] Terminal: Converts voice input to a digital format and saves it locally.

[0593] Terminal: Sends stored digital audio data to a cloud-based speech recognition engine.

[0594] Server: The speech recognition engine converts the digital speech data into text data "Hello, what is your name?".

[0595] 3. Text Translation

[0596] Server: Sends the speech-recognized text to the translation engine.

[0597] Server: The translation engine translates the text into English as "Hello, what is your name?".

[0598] 4. Speech synthesis

[0599] Server: Sends the translated English text "Hello, what is your name?" to the speech synthesis engine.

[0600] Server: The speech synthesis engine converts English text into speech data.

[0601] Server: Sends the generated audio data to the terminal.

[0602] 5. Audio Playback

[0603] Terminal: Receives audio data and plays it back to the user.

[0604] Specific example

[0605] For example, the specific process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0606] 1. User: "Hello, what's your name?" they say to their smartphone.

[0607] 2. Device: Record audio, convert it to a digital format, and save it.

[0608] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[0609] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[0610] 5. Server: The translation engine translates the text into English as "Hello, what is your name?".

[0611] 6. Server: The speech synthesis engine converts English text into speech data.

[0612] 7. Server: Sends audio data to the user's terminal.

[0613] 8. Terminal: Play the audio data "Hello, what is your name?" to the user.

[0614] 9. The other person responds, "My name is John."

[0615] 10. The other party's device: Record the audio, convert it to a digital format, and save it.

[0616] 11. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[0617] 12. Server: The speech recognition engine converts "My name is John." into text.

[0618] 13. Server: The translation engine translates the text into Japanese as "My name is John."

[0619] 14. Server: The speech synthesis engine converts Japanese text into speech data.

[0620] 15. Server: Sends audio data to the user's terminal.

[0621] 16. Terminal: Play the audio data "My name is John." to the user.

[0622] The above process enables smooth communication between users and people who speak a foreign language.

[0623] The following describes the processing flow.

[0624] Step 1:

[0625] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[0626] Step 2:

[0627] Terminal: The voice input application records the user's speech and converts it into a digital format.

[0628] Step 3:

[0629] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[0630] Step 4:

[0631] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[0632] Step 5:

[0633] Server: Sends the speech-recognized text data to the translation engine.

[0634] Step 6:

[0635] Server: The translation engine translates the text data into English as "Hello, what is your name?".

[0636] Step 7:

[0637] Server: Sends the translated English text data to the speech synthesis engine.

[0638] Step 8:

[0639] Server: The speech synthesis engine converts English text data into speech data.

[0640] Step 9:

[0641] Server: Sends the generated audio data to the terminal.

[0642] Step 10:

[0643] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio "Hello, what is your name?".

[0644] Step 11:

[0645] The other party (device): The other party responds with "My name is John."

[0646] Step 12:

[0647] Recipient's device: Records the recipient's speech and converts it to a digital format.

[0648] Step 13:

[0649] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[0650] Step 14:

[0651] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[0652] Step 15:

[0653] Server: Sends the speech-recognized text data to the translation engine.

[0654] Step 16:

[0655] Server: The translation engine translates the text data into Japanese as "My name is John."

[0656] Step 17:

[0657] Server: Sends the translated Japanese text data to the speech synthesis engine.

[0658] Step 18:

[0659] Server: The speech synthesis engine converts Japanese text data into speech data.

[0660] Step 19:

[0661] Server: Sends the generated audio data to the user's device.

[0662] Step 20:

[0663] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio saying, "My name is John."

[0664] This series of steps enables real-time voice translation between the user and the person speaking the foreign language, facilitating smooth communication.

[0665] (Example 1)

[0666] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0667] In modern society, smooth communication between people who speak different languages ​​is a challenging task. In particular, there is a lack of technology for real-time translation and seamless multilingual conversation. Therefore, there is a need to develop systems that allow users who speak different languages ​​to communicate easily.

[0668] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0669] In this invention, the server includes means for inputting the user's voice and converting the voice into a digital format; speech recognition means for converting the digitally converted voice into text; translation means for translating the text into a predetermined foreign language; speech synthesis means for converting the translated text into audio data; playback means for playing the audio data to the user; means for transmitting the text data obtained from the speech recognition means to a translation engine; means for transmitting the translated text obtained from the translation engine to a speech synthesis engine; and means for transmitting the audio data obtained from the speech synthesis engine to the user's terminal. This enables users who speak different languages ​​to communicate smoothly in real time.

[0670] A "user" refers to a person who uses a system to communicate.

[0671] "Means of voice input" refers to devices and technologies for converting a user's voice into a digital format.

[0672] "Digital format" refers to a state in which audio or data has been converted into a format that can be processed by electronic devices.

[0673] "Speech recognition means" refers to a function or technology that converts speech, which has been converted into a digital format, into text.

[0674] "Translation means" refers to a function or technology for translating text data into a specified foreign language.

[0675] "Speech synthesis means" refers to a function or technology for converting text data translated into a foreign language into speech data.

[0676] "Playback means" refers to devices and technologies for providing audio data to users as auditory information.

[0677] A "cloud-based speech recognition engine" refers to a speech recognition service provided via the internet.

[0678] "User's personal information terminal" refers to portable electronic devices such as smartphones and tablets.

[0679] "Audio output device" refers to a device that includes speakers or headphones for playing back audio data.

[0680] "Text data" refers to standard string information converted by speech recognition.

[0681] "Translation" refers to the text data after it has been translated.

[0682] "Audio data" refers to the format of digital audio generated by speech synthesis technology.

[0683] This invention is a system aimed at enabling users who speak different languages ​​to communicate smoothly in real time. The system includes a series of processes that convert audio into a digital format, convert it to text, translate it, convert it back into audio data, and finally play it back to the user.

[0684] System Configuration

[0685] This system consists of the following main elements:

[0686] 1. Voice input method

[0687] 2. Speech recognition means

[0688] 3. Translation methods

[0689] 4. Speech synthesis means

[0690] 5. Regeneration means

[0691] Hardware and software to use

[0692] For voice input, the user's mobile device, such as a smartphone or tablet, is used. These devices have built-in microphones that record the user's voice.

[0693] The speech recognition method sends recorded audio to a cloud-based speech recognition engine, which then converts the audio to text. Specifically, it uses APIs such as the Google Cloud Speech-to-Text API.

[0694] The translation method translates the text obtained by the speech recognition method into a specified foreign language. This uses a translation engine such as the Google Cloud Translation API.

[0695] The speech synthesis method converts translated text into speech data. A specific example used is the Microsoft Azure Text-to-Speech API.

[0696] The playback method involves playing the generated audio data to the user. The speaker or earphones built into the user's smartphone or tablet are used as the playback device.

[0697] Specific example

[0698] For example, the process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0699] 1. The user says to their smartphone, "Hello, what's your name?"

[0700] 2. The device records the audio, converts it to a digital format, and saves it.

[0701] 3. The device sends the digital voice data to a cloud-based speech recognition engine (Google Cloud Speech-to-Text API).

[0702] 4. The server converts the audio data into text data.

[0703] 5. The server sends the text data to the translation engine (Google Cloud Translation API) for translation into English.

[0704] 6. The server sends the English text data to the text-to-speech engine (Microsoft Azure Text-to-Speech API) and converts it into speech data.

[0705] 7. The server sends the generated English audio data to the user's device.

[0706] 8. The device plays audio data, and an English voice says, "Hello, what is your name?"

[0707] Example of a prompt

[0708] The following are specific examples of prompt statements used in this system.

[0709] Prompt: Translate the Japanese phrase "Hello, what is your name?" into English, and then generate the English audio.

[0710] Using this prompt, the system goes through the processes of speech recognition, text translation, and speech synthesis to generate and provide results to the user.

[0711] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0712] Step 1:

[0713] Voice input

[0714] The user speaks into their smartphone or tablet.

[0715] Input: User's voice (e.g., "Hello, what's your name?")

[0716] Specific action: The user speaks into the microphone.

[0717] Output: Raw audio data

[0718] Step 2:

[0719] Audio recording and digital conversion

[0720] The device records the user's voice and converts it to a digital format.

[0721] Input: Raw audio data

[0722] Specific operation: The smartphone's audio engine (e.g., AudioRecord API) converts the raw audio into PCM digital data.

[0723] Output: Digital audio data in PCM format

[0724] Step 3:

[0725] Sending audio data

[0726] The device sends the converted digital audio data to a cloud-based speech recognition engine.

[0727] Input: PCM format digital audio data

[0728] Specific operation: Send digital audio data as an HTTP request to the Google Cloud Speech-to-Text API.

[0729] Output: Audio data sent to the server

[0730] Step 4:

[0731] Speech recognition

[0732] The server uses a speech recognition engine to convert digital audio data into text data.

[0733] Input: PCM format digital audio data

[0734] Specific operation: The Google Cloud Speech-to-Text API analyzes the audio data and converts it into text, "Hello, what is your name?".

[0735] Output: Text data "Hello, what is your name?"

[0736] Step 5:

[0737] Text translation

[0738] The server sends the text data acquired through speech recognition to the translation engine.

[0739] Input: Text data "Hello, what is your name?"

[0740] Specific operation: The server sends a translation request to the Google Cloud Translation API. The request includes the source text and the target language.

[0741] Output: English translation text data "Hello, what is your name?"

[0742] Step 6:

[0743] Preparation for speech synthesis

[0744] The server sends the translated English text to the speech synthesis engine.

[0745] Input: English text data "Hello, what is your name?"

[0746] Specific operation: The server sends a text-to-speech request to the Microsoft Azure Text-to-Speech API. The request includes the translated text and voice configuration information.

[0747] Output: Voice data request

[0748] Step 7:

[0749] Speech synthesis

[0750] The server uses a speech synthesis engine to convert text data into speech data.

[0751] Input: English text data "Hello, what is your name?"

[0752] Specific operation: The Microsoft Azure Text-to-Speech API parses English text and converts it into speech data.

[0753] Output: English audio data

[0754] Step 8:

[0755] Sending audio data

[0756] The server sends the generated English audio data to the user's device.

[0757] Input: English audio data

[0758] Specific operation: The server sends the generated audio data to the terminal as an HTTP response.

[0759] Output: Audio data sent to the user's terminal

[0760] Step 9:

[0761] Playback of audio data

[0762] The device plays the received audio data.

[0763] Input: English audio data

[0764] Specific action: The device's media player (e.g., MediaPlayer API) plays audio data, and the voice "Hello, what is your name?" is played through the speaker.

[0765] Output: Audio played to the user

[0766] (Application Example 1)

[0767] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0768] In brick-and-mortar retail environments, where smooth communication between customers and employees with different language backgrounds is essential, language barriers can prevent customers from obtaining necessary information, leading to a decline in service quality. Furthermore, the difficulty for employees to be fluent in multiple languages ​​reduces operational efficiency in physical stores. A system is needed to address these challenges and support communication in brick-and-mortar retail settings.

[0769] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0770] In this invention, the server includes means for inputting a user's voice and converting the voice into a digital format; voice recognition means for converting the voice converted into digital format into text; translation means for translating the text into a predetermined foreign language; voice synthesis means for converting the text translated into the foreign language into audio data; playback means for playing the audio data back to the user; means for supporting communication between customers and employees in a store; and means for customers or employees to speak into the device. This enables real-time voice translation and smooth communication between customers and employees who speak different languages.

[0771] "User" refers to the customers and employees who use this system.

[0772] "Means for inputting audio and converting said audio into a digital format" refers to a device or software that has the function of converting analog audio data into digital data.

[0773] "Speech recognition means" refers to a device or software that has the function of converting speech converted into digital format into text format.

[0774] "Translation means" refers to a device or software that has the function of translating text, which has been converted into text format by speech recognition means, into a predetermined foreign language.

[0775] "Speech synthesis means" refers to a device or software that has the function of converting translated text into speech data.

[0776] "Playback means" refers to a device or software that has the function of playing back audio data generated by a speech synthesis means to the user.

[0777] "Means for supporting communication between customers and employees within a store" refers to devices or software that provide the necessary functions for smooth communication between customers and employees who speak different languages ​​in a physical store.

[0778] "Device" refers to a personal information terminal (such as a smartphone or tablet) used for inputting and playing audio.

[0779] This invention is a multilingual real-time interpretation assistant system aimed at facilitating smooth communication between customers and employees who speak different languages. This invention is primarily implemented using mobile devices such as smartphones and tablets, and utilizes a cloud-based speech recognition engine and translation engine.

[0780] System Configuration

[0781] This system includes the following key hardware and software components:

[0782] 1. Mobile information terminal

[0783] It has a microphone and processor for voice input, recording, and conversion to digital format.

[0784] 2. Cloud-based speech recognition engine

[0785] Use the Google Cloud Speech-to-Text API to convert digital audio data into text data.

[0786] 3. Translation engine

[0787] The retrieved text data is translated into a specified foreign language using the Google Cloud Translate API.

[0788] 4. Speech synthesis engine

[0789] The Google Cloud Text-to-Speech API is used to convert translated text data into audio data.

[0790] 5. Regeneration means

[0791] Audio data is played using the speaker installed in the mobile device.

[0792] Processing flow details

[0793] The server includes means for inputting user voice and converting said voice into a digital format, voice recognition means, translation means, speech synthesis means, and playback means. By coordinating these means, real-time communication between customers and employees in a physical store is realized.

[0794] 1. Voice input and conversion

[0795] Personal digital assistant (PDA): The user (customer or employee) speaks into the device. The device records the audio and converts it to a digital format.

[0796] 2. Speech Recognition

[0797] Mobile device: Sends digital audio data to the Google Cloud Speech-to-Text API. The Google Cloud Speech-to-Text API converts the audio data into text data.

[0798] 3. Text Translation

[0799] Server: Sends text data to the Google Cloud Translate API for translation into the specified foreign language.

[0800] 4. Speech synthesis

[0801] Server: Sends the translated text data to the Google Cloud Text-to-Speech API and converts it into speech data.

[0802] 5. Audio Playback

[0803] Personal digital assistant (PDCA): Receives generated audio data and plays it through the speaker.

[0804] Specific example

[0805] For example, consider a case where a Japanese-speaking customer asks an English-speaking employee for the location of a product.

[0806] 1. Customer: "Where can I find this product?" they say to their smartphone.

[0807] 2. Personal digital assistant (PDCA): Records audio, converts it to a digital format, and sends it to the cloud.

[0808] 3. Google Cloud Speech-to-Text API: Converts audio data into text data such as "Where can I find this product?".

[0809] 4. Google Cloud Translate API: Translate the text data into "Where is this product located?".

[0810] 5. Google Cloud Text-to-Speech API: Converts translated text data into speech data.

[0811] 6. Mobile device: Play the English audio "Where is this product located?".

[0812] Example of a prompt

[0813] "Where can I find this product?"

[0814] In this way, customers and employees can communicate smoothly, overcoming language barriers.

[0815] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0816] Step 1:

[0817] Voice input and conversion to digital format

[0818] User: The user (customer or employee) speaks into a mobile device.

[0819] Input: Analog audio data.

[0820] Terminal: The microphone of the personal digital assistant (PDCA) records audio, and the analog audio data is converted into a digital format.

[0821] Output: Digital audio data.

[0822] Step 2:

[0823] Speech recognition of digital audio data

[0824] Terminal: Sends digital audio data to the cloud.

[0825] Input: Digital audio data.

[0826] Server (Google Cloud Speech-to-Text API): Receives digital audio data and performs speech recognition. Specifically, it analyzes audio patterns and converts them into text data.

[0827] Output: Text data (e.g., "Where can I find this product?").

[0828] Step 3:

[0829] Text data translation

[0830] Server: The translation engine (Google Cloud Translate API) translates text data into the specified foreign language.

[0831] Input: Japanese text data (e.g., "Where can I find this product?").

[0832] Server (Google Cloud Translate API): Receives text data and performs translation processing. Based on the model, it translates Japanese into the target language, such as English.

[0833] Output: Translated text data (e.g., "Where is this product located?").

[0834] Step 4:

[0835] Converting translated text to audio data

[0836] Server: Sends the translated text data to the text-to-speech engine (Google Cloud Text-to-Speech API) and converts it into speech data.

[0837] Input: Translated text data (e.g., "Where is this product located?").

[0838] Server (Google Cloud Text-to-Speech API): Converts text data into speech data. Specifically, it analyzes the text, breaks it down into phonemes, and then synthesizes them together.

[0839] Output: Audio data (e.g., "Where is this product located?").

[0840] Step 5:

[0841] Playback of audio data

[0842] Terminal: Receives audio data and plays it back to the user.

[0843] Input: Audio data received from the server (e.g., "Where is this product located?").

[0844] Device (speaker): Plays audio data.

[0845] Output: Voice response message (e.g., "Where is this product located?").

[0846] This processing step enables users to interact in real time with people who speak different languages.

[0847] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0848] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. It also provides a system that combines this with an emotion engine that recognizes the user's emotions and appropriately adjusts the translated expression accordingly. The invention will describe the system's program processing in natural language, and will also include specific usage examples.

[0849] System Overview

[0850] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system includes the following main elements:

[0851] 1. Voice input method

[0852] 2. Speech recognition means

[0853] 3. Translation methods

[0854] 4. Emotional Engine

[0855] 5. Speech synthesis means

[0856] 6. Reproduction means

[0857] System processing flow

[0858] 1. Voice input

[0859] User (terminal): The user speaks towards a smartphone or tablet. For example, make a statement like "Hello, what's your name?"

[0860] 2. Speech recognition

[0861] Terminal: Convert the voice input into digital format and save it locally.

[0862] Terminal: Send the saved digital voice data to a cloud-based speech recognition API.

[0863] Server: The speech recognition engine converts the digital voice data into text data "Hello, what's your name?"

[0864] 3. Emotion recognition

[0865] Server: The emotion engine analyzes the emotion from the user's voice in real time and generates emotion data. For example, recognize that the statement "Hello, what's your name?" is in a happy tone.

[0866] 4. Text translation

[0867] Server: Send the speech-recognized text data and emotion data to the translation engine.

[0868] Server: The translation engine translates the text data "Hello, what's your name?" into English as "Hello, what is your name?" and adjusts the expression to fit the user's emotion.

[0869] 5. Speech synthesis

[0870] Server: Sends the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[0871] Server: The speech synthesis engine converts English text data into speech data.

[0872] Server: Sends the generated audio data to the terminal.

[0873] 6. Audio Playback

[0874] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play a cheerful voice saying, "Hello, what is your name?"

[0875] Specific example

[0876] For example, the processing flow when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[0877] 1. User: "Hello, what's your name?" they say to their smartphone.

[0878] 2. Device: Record audio, convert it to a digital format, and save it.

[0879] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[0880] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[0881] 5. Server: The emotion engine recognizes emotions from the user's voice and generates "happy" emotion data.

[0882] 6. Server: The translation engine translates the text into English as "Hello, what is your name?", maintaining the cheerful tone.

[0883] 7. Server: The speech synthesis engine converts English text into speech data.

[0884] 8. Server: Sends audio data to the user's terminal.

[0885] 9. Device: Play the voice data "Hello, what is your name?" in a cheerful tone.

[0886] 10. The other person responds, "My name is John."

[0887] 11. The other party's device: Record the audio, convert it to a digital format, and save it.

[0888] 12. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[0889] 13. Server: The speech recognition engine converts "My name is John." into text.

[0890] 14. Server: The emotion engine recognizes emotions from the other party's voice and generates "calm" emotion data.

[0891] 15. Server: The translation engine translates the text into Japanese as "My name is John," maintaining a calm tone.

[0892] 16. Server: The speech synthesis engine converts Japanese text into speech data.

[0893] 17. Server: Sends audio data to the user's terminal.

[0894] 18. Device: Play the audio data "My name is John." in a calm tone.

[0895] In this way, smooth and considerate communication between the user and the person speaking the foreign language is achieved.

[0896] The following describes the processing flow.

[0897] Step 1:

[0898] User (device): The user speaks to their smartphone or tablet and says, "Hello, what's your name?"

[0899] Step 2:

[0900] Terminal: The voice input application records the user's speech and converts it into a digital format.

[0901] Step 3:

[0902] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[0903] Step 4:

[0904] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[0905] Step 5:

[0906] Server: Sends text data to the emotion engine to analyze the user's emotions.

[0907] Step 6:

[0908] Server: The emotion engine analyzes the tone and pitch of the voice, etc., and recognizes that the user has a "happy" emotion. It also generates emotion data.

[0909] Step 7:

[0910] Server: Send the voice-recognized text data and emotion data to the translation engine.

[0911] Step 8:

[0912] Server: The translation engine translates the text data "こんにちは、お名前は何ですか?" to "Hello, what is your name?" in English and adjusts the expression to fit the user's "happy" emotion.

[0913] Step 9:

[0914] Server: Send the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[0915] Step 10:

[0916] Server: The speech synthesis engine converts the English text data into voice data.

[0917] Step 11:

[0918] Server: Send the generated voice data to the terminal.

[0919] Step 12:

[0920] Terminal: Receive the voice data and play it as voice to the user. For example, the voice "Hello, what is your name?" is played in a happy tone.

[0921] Step 13:

[0922] The other party (device): The other party responds with "My name is John."

[0923] Step 14:

[0924] Recipient's device: Records the recipient's speech and converts it to a digital format.

[0925] Step 15:

[0926] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[0927] Step 16:

[0928] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[0929] Step 17:

[0930] Server: Sends text data to the emotion engine to analyze the other party's emotions.

[0931] Step 18:

[0932] Server: The emotion engine analyzes the tone and pitch of the voice and recognizes that the other party has a "calm" emotion. It also generates emotion data.

[0933] Step 19:

[0934] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[0935] Step 20:

[0936] Server: The translation engine translates the text data "My name is John." into Japanese as "私の名前はジョンです。" and adjusts the expression to fit the "calm" emotions of the other party.

[0937] Step 21:

[0938] Server: The translated Japanese text data "私の名前はジョンです。" is sent to the text-to-speech engine.

[0939] Step 22:

[0940] Server: The text-to-speech engine converts the Japanese text data into audio data.

[0941] Step 23:

[0942] Server: The generated audio data is sent to the user's terminal.

[0943] Step 24:

[0944] Terminal: Receives the audio data and plays it as audio for the user. For example, the audio is played in a calm tone as "私の名前はジョンです。".

[0945] In this way, smooth communication considering emotions is realized between the user and the foreign language speaker.

[0946] (Example 2)

[0947] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0948] In recent years, international communication has increased, and smooth multilingual conversation is required. However, conventional translation systems simply translate languages ​​without adjusting the translation to take into account the speaker's emotions, resulting in a decline in the quality of communication. Furthermore, in real-time speech translation, the processing speed from speech recognition to translation and playback has been a problem. Therefore, there is a need for a system that accurately reflects the speaker's emotions while enabling smooth, real-time communication between multiple languages.

[0949] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0950] In this invention, the server includes means for recognizing emotions from the user's voice in real time and generating emotion data, means for adjusting translated expressions based on translated text and emotion data, and means for converting text translated into a foreign language into speech data. This enables real-time multilingual voice communication using translated expressions that reflect the speaker's emotions.

[0951] A "user" refers to a person who uses the system to perform voice input and voice translation.

[0952] "Digital format" refers to a format in which analog audio is converted into digital data that can be processed by a computer.

[0953] "Speech recognition means" refers to means that have the function of converting digital speech data into text data.

[0954] "Translation means" refers to a means that has the function of converting text from one language to another language.

[0955] "Emotion recognition means" refers to means that have the function of analyzing the speaker's emotions in real time from audio data and generating emotion data.

[0956] "Means for adjusting translated expressions" refers to means that have the function of adjusting translated expressions to reflect appropriate sentiment based on the translated text and sentiment data.

[0957] "Speech synthesis means" refers to means that have the function of converting text data into speech data.

[0958] "Playback means" refers to means that have the function of outputting the generated audio data as sound.

[0959] A "cloud-based speech recognition engine" refers to an engine that uses servers and software accessible via a network to convert speech data into text data.

[0960] "Communication equipment" refers to electronic devices with communication capabilities, such as smartphones and tablets.

[0961] "Audio equipment" refers to devices such as speakers and earphones used to play audio data.

[0962] This invention is a speech translation system that enables users to communicate smoothly with people who speak a foreign language. The system consists of the following processes: speech input, speech recognition, emotion recognition, text translation, speech synthesis, and speech playback. Each process utilizes specific hardware and software.

[0963] System Configuration

[0964] 1. Voice input

[0965] User: Speaks into a smartphone or tablet. For example, says, "Hello, what's your name?"

[0966] Terminal: Uses a microphone as an interface to convert the user's voice into a digital format.

[0967] 2. Speech Recognition

[0968] Terminal: Sends digital voice data to a cloud-based speech recognition engine. The API used is the speech recognition engine.

[0969] Server: Converts the digital audio data received by the speech recognition engine into text data.

[0970] 3. Emotion recognition

[0971] Server: The emotion recognition engine analyzes the user's voice in real time and generates emotion data. The emotion recognition engine used is an emotion analysis engine.

[0972] 4. Text Translation

[0973] Server: Sends the speech-recognized text data and generated sentiment data to the translation engine. The translation engine used is a text translation engine.

[0974] Server: The translation engine converts the text into the target language and adjusts the translation based on sentiment data.

[0975] 5. Speech synthesis

[0976] Server: Sends the translated text data to the speech synthesis engine. The speech synthesis engine used is speech synthesis software.

[0977] Server: The speech synthesis engine converts text data into speech data.

[0978] 6. Audio Playback

[0979] Terminal: Receives audio data and plays it back using the terminal's sound device. For example, it might play back "Hello, what is your name?" in a cheerful tone.

[0980] Specific example

[0981] As a concrete example, we will simulate a conversation between a Japanese-speaking user and an English-speaking partner. The following is the processing flow.

[0982] 1. User: "Hello, what's your name?" they say to their smartphone.

[0983] 2. Device: Records audio and converts it into digital data.

[0984] 3. Server: The speech recognition engine converts "Hello, what is your name?" into text data.

[0985] 4. Server: The emotion recognition engine generates text data and emotion data, and detects the emotion of "happiness".

[0986] 5. Server: The translation engine translates the text data into English as "Hello, what is your name?", maintaining a cheerful tone.

[0987] 6. Server: The speech synthesis engine converts the translated text into speech data.

[0988] 7. Terminal: Receives audio data and plays "Hello, what is your name?" in a cheerful tone.

[0989] Example of a prompt

[0990] Use the following prompts to input data into the generative AI model.

[0991] "Please translate the Japanese phrase 'Hello, what is your name?' into English and convert it into speech, while maintaining the emotion in the speech."

[0992] "Translate the English sentence 'My name is John.' into Japanese and then speak it in a calm tone."

[0993] In this way, real-time voice translation that takes emotions into consideration is achieved between the user and the person speaking the foreign language.

[0994] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0995] Program processing flow

[0996] Step 1: Voice Input

[0997] User: Speak into the microphone on your smartphone or tablet and say, "Hello, what's your name?"

[0998] Input: Analog audio

[0999] Output: Digital audio data

[1000] Terminal: Captures the user's voice with a microphone, converts the analog audio to a digital format using an audio digitization module, and saves it to a buffer.

[1001] Step 2: Speech Recognition

[1002] Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[1003] Input: Digital audio data

[1004] Output: Text data ("Hello, what is your name?")

[1005] Server: The speech recognition engine analyzes the digital speech data it receives and converts it into text data, taking into account phonemes and context.

[1006] Step 3: Emotion Recognition

[1007] Server: Sends the speech-recognized text data and digital audio data to the emotion recognition engine.

[1008] Input: Text data, digital audio data

[1009] Output: Emotional data (e.g., "Happy")

[1010] Server: The emotion recognition engine analyzes the tone, pitch, and speed of the voice, evaluates the user's emotional state in real time, and generates emotion data.

[1011] Step 4: Text Translation

[1012] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[1013] Input: Text data, sentiment data

[1014] Output: Translated text data (e.g., "Hello, what is your name?")

[1015] Server: The translation engine converts text data into the target foreign language and adjusts the translation based on sentiment data.

[1016] Step 5: Speech Synthesis

[1017] Server: Sends the translated text data to the speech synthesis engine.

[1018] Input: Translated text data

[1019] Output: Audio data

[1020] Server: The speech synthesis engine analyzes the translated text data and generates speech data that reflects emotional tone.

[1021] Step 6: Audio Playback

[1022] Device: Receives audio data and plays it through the device's speaker.

[1023] Input: Audio data

[1024] Output: Played audio (e.g., "Hello, what is your name?")

[1025] Terminal: Uses an audio device to play the generated audio data to the user.

[1026] (Application Example 2)

[1027] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1028] In physical stores, communication between customers who speak different languages ​​and employees often breaks down. This can lead to decreased customer satisfaction and a decline in the quality of service. Furthermore, it is difficult for employees to immediately understand and respond to multiple languages, especially when it comes to conveying nuances, including emotions. In this situation, it is necessary to improve customer service in physical stores by streamlining foreign language support and providing accurate translations that take emotions into consideration.

[1029] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1030] This invention includes a server that includes means for inputting a user's voice and converting the voice into a digital format, speech recognition means for converting the digitally converted voice into text, translation means for translating the text into a predetermined foreign language, an emotion engine that recognizes the emotion of the translated text and adjusts the translated expression appropriately according to the emotion, speech synthesis means for converting the translated text into audio data, playback means for playing the audio data back to the user, and a device that implements an application that supports real-time multilingual communication between the user and customers who speak a foreign language. This makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the emotions of customers.

[1031] "Means of inputting audio and converting it to a digital format" refers to devices or processes that receive audio signals and record them as digital data.

[1032] "Speech recognition means" refers to a technology or device that analyzes speech converted into digital data and converts it into corresponding text data.

[1033] "Translation methods" refer to software or algorithms used to convert text data from one language into another language.

[1034] An "emotion engine" is a technology or algorithm that recognizes a speaker's emotions from audio or text data and analyzes that emotional information.

[1035] "Speech synthesis means" refers to a technology or device that analyzes text data and generates speech data corresponding to that text.

[1036] "Reproduction means" refers to a device or technology used to actually pronounce the generated audio data as sound.

[1037] "Device" is a general term for any equipment on which an application is installed, such as communication terminals, smartphones, and smart glasses.

[1038] A "cloud-based speech recognition engine" refers to a service or technology for performing speech recognition in a cloud computing environment.

[1039] A "communication terminal" refers to an electronic device such as a smartphone or tablet that has functions such as voice input and playback, and data communication.

[1040] An "application that supports multilingual communication" is software designed to enable users who speak different languages ​​to communicate with each other in real time.

[1041] This invention is a system that facilitates multilingual support in physical stores and provides appropriate translations that take emotions into consideration. This system uses a device to support real-time communication between users and foreign language-speaking customers. The main elements of the system are means for inputting voice and converting it to a digital format, voice recognition means, translation means, emotion engine, voice synthesis means, playback means, and a device for implementing the application.

[1042] The following hardware and software will be used to implement this system:

[1043] Hardware: Communication devices (e.g., smartphones, tablets, smart glasses)

[1044] software:

[1045] Speech recognition: Google Cloud Speech-to-Text API

[1046] Translation: Google Cloud Translation API

[1047] Emotion recognition: IBM Watson Tone Analyzer API

[1048] Text-to-speech: Google Cloud Text-to-Speech API

[1049] Real-time communication: Firebase Realtime Database

[1050] Description of the system's program processing

[1051] First, the user speaks into a communication device. The communication device converts this audio into a digital format and uploads it to Firebase. The server sends the audio data stored in Firebase to the Google Cloud Speech-to-Text API, where it converts the audio into text data.

[1052] Next, the server sends this text data to the IBM Watson Tone Analyzer API for sentiment analysis. The text, including sentiment data, is then sent to the Google Cloud Translation API for translation into the specified foreign language. Based on the sentiment data, the translated sentence is adjusted to an appropriate tone.

[1053] The server then sends the translated text data to the Google Cloud Text-to-Speech API to generate audio data. This audio data is then stored again in Firebase and sent to the user's communication device. The communication device plays this audio data, allowing the user to provide the translated message to their customers.

[1054] Specific example

[1055] Let's consider a situation where an employee working in a souvenir shop in a tourist area needs to communicate smoothly with foreign customers. For example,

[1056] User (employee): "Welcome, how can I help you?" they say to their smartphone.

[1057] System: The server recognizes the speech, analyzes the emotions, translates it into English, and plays back "Welcome! How can I help you?" in a friendly tone.

[1058] The other party (customer) asks, "Can you tell me where I can find local souvenirs?"

[1059] User: Provide a properly translated response through the system again to facilitate smooth communication.

[1060] Example of a prompt

[1061] "Based on the following information, please generate code that performs translations tailored to the customer's emotions."

[1062] Customer comment: "Hello, can I purchase this?"

[1063] Emotion: "Friendly"

[1064] Translated statement: "Hello, can I purchase this?"

[1065] Code for speech synthesis in a friendly tone

[1066] This invention makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the customer's feelings.

[1067] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1068] Step 1:

[1069] Input: The user speaks into the communication terminal.

[1070] Operation: The user speaks to the communication terminal saying, "Welcome, how can I help you?"

[1071] Data processing: The communication terminal converts this audio into a digital format.

[1072] Output: Digital audio data is generated.

[1073] Step 2:

[1074] Input: Digital audio data.

[1075] Operation: The communication device uploads this digital audio data to Firebase.

[1076] Data processing: Digital audio data is transferred to the cloud.

[1077] Output: Digital audio data is stored on Firebase.

[1078] Step 3:

[1079] Input: Digital audio data stored in Firebase.

[1080] Operation: The server retrieves audio data from Firebase and sends it to the Google Cloud Speech-to-Text API.

[1081] Data processing: The speech recognition engine analyzes the digital audio data and converts it into text data.

[1082] Output: The text data "Welcome, how can I help you?" is generated.

[1083] Step 4:

[1084] Input: Text data.

[1085] Operation: The server sends text data to the IBM Watson Tone Analyzer API.

[1086] Data processing: The emotion engine analyzes the text data and generates emotion data (e.g., "friendly tone").

[1087] Output: Sentiment data is generated and attached to the text data.

[1088] Step 5:

[1089] Input: Text data and sentiment data.

[1090] Operation: The server sends this data to the Google Cloud Translation API.

[1091] Data processing: The translation engine translates text data into the specified foreign language and adjusts it based on sentiment data.

[1092] Output: The translated text data "Welcome! How can I help you?" is generated.

[1093] Step 6:

[1094] Input: Translated text data.

[1095] Operation: The server sends this text data to the Google Cloud Text-to-Speech API.

[1096] Data processing: The speech synthesis engine converts the translated text data into speech data.

[1097] Output: The audio data "Welcome! How can I help you?" is generated.

[1098] Step 7:

[1099] Input: Translated audio data.

[1100] Operation: The server uploads the generated audio data to Firebase and sends it to the user's communication device.

[1101] Data processing: Audio data is transferred from the server to the communication terminal.

[1102] Output: The audio data is saved to the communication terminal.

[1103] Step 8:

[1104] Input: Translated audio data.

[1105] Operation: The communication terminal plays the translated audio data through its speaker.

[1106] Data processing: Audio data is output as audio.

[1107] Output: The translated audio "Welcome! How can I help you?" is played to the customer.

[1108] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1109] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1110] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1111] [Third Embodiment]

[1112] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1113] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1114] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1115] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1116] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1117] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1118] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1119] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1120] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1121] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1122] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1123] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1124] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. The system's program processing will be described in natural language to illustrate embodiments of the invention. Specific usage examples will also be provided.

[1125] System Overview

[1126] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system consists of the following main elements:

[1127] 1. Voice input method

[1128] 2. Speech recognition means

[1129] 3. Translation methods

[1130] 4. Speech synthesis means

[1131] 5. Regeneration means

[1132] System processing flow

[1133] 1. Voice input

[1134] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[1135] 2. Speech Recognition

[1136] Terminal: Converts voice input to a digital format and saves it locally.

[1137] Terminal: Sends stored digital audio data to a cloud-based speech recognition engine.

[1138] Server: The speech recognition engine converts the digital speech data into text data "Hello, what is your name?".

[1139] 3. Text Translation

[1140] Server: Sends the speech-recognized text to the translation engine.

[1141] Server: The translation engine translates the text into English as "Hello, what is your name?".

[1142] 4. Speech synthesis

[1143] Server: Sends the translated English text "Hello, what is your name?" to the speech synthesis engine.

[1144] Server: The speech synthesis engine converts English text into speech data.

[1145] Server: Sends the generated audio data to the terminal.

[1146] 5. Audio Playback

[1147] Terminal: Receives audio data and plays it back to the user.

[1148] Specific example

[1149] For example, the specific process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[1150] 1. User: "Hello, what's your name?" they say to their smartphone.

[1151] 2. Device: Record audio, convert it to a digital format, and save it.

[1152] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[1153] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[1154] 5. Server: The translation engine translates the text into English as "Hello, what is your name?".

[1155] 6. Server: The speech synthesis engine converts English text into speech data.

[1156] 7. Server: Sends audio data to the user's terminal.

[1157] 8. Terminal: Play the audio data "Hello, what is your name?" to the user.

[1158] 9. The other person responds, "My name is John."

[1159] 10. The other party's device: Record the audio, convert it to a digital format, and save it.

[1160] 11. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[1161] 12. Server: The speech recognition engine converts "My name is John." into text.

[1162] 13. Server: The translation engine translates the text into Japanese as "My name is John."

[1163] 14. Server: The speech synthesis engine converts Japanese text into speech data.

[1164] 15. Server: Sends audio data to the user's terminal.

[1165] 16. Terminal: Play the audio data "My name is John." to the user.

[1166] The above process enables smooth communication between users and people who speak a foreign language.

[1167] The following describes the processing flow.

[1168] Step 1:

[1169] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[1170] Step 2:

[1171] Terminal: The voice input application records the user's speech and converts it into a digital format.

[1172] Step 3:

[1173] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[1174] Step 4:

[1175] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[1176] Step 5:

[1177] Server: Sends the speech-recognized text data to the translation engine.

[1178] Step 6:

[1179] Server: The translation engine translates the text data into English as "Hello, what is your name?".

[1180] Step 7:

[1181] Server: Sends the translated English text data to the speech synthesis engine.

[1182] Step 8:

[1183] Server: The speech synthesis engine converts English text data into speech data.

[1184] Step 9:

[1185] Server: Sends the generated audio data to the terminal.

[1186] Step 10:

[1187] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio "Hello, what is your name?".

[1188] Step 11:

[1189] The other party (device): The other party responds with "My name is John."

[1190] Step 12:

[1191] Recipient's device: Records the recipient's speech and converts it to a digital format.

[1192] Step 13:

[1193] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[1194] Step 14:

[1195] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[1196] Step 15:

[1197] Server: Sends the speech-recognized text data to the translation engine.

[1198] Step 16:

[1199] Server: The translation engine translates the text data into Japanese as "My name is John."

[1200] Step 17:

[1201] Server: Sends the translated Japanese text data to the speech synthesis engine.

[1202] Step 18:

[1203] Server: The speech synthesis engine converts Japanese text data into speech data.

[1204] Step 19:

[1205] Server: Sends the generated audio data to the user's device.

[1206] Step 20:

[1207] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio saying, "My name is John."

[1208] This series of steps enables real-time voice translation between the user and the person speaking the foreign language, facilitating smooth communication.

[1209] (Example 1)

[1210] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1211] In modern society, smooth communication between people who speak different languages ​​is a challenging task. In particular, there is a lack of technology for real-time translation and seamless multilingual conversation. Therefore, there is a need to develop systems that allow users who speak different languages ​​to communicate easily.

[1212] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1213] In this invention, the server includes means for inputting the user's voice and converting the voice into a digital format; speech recognition means for converting the digitally converted voice into text; translation means for translating the text into a predetermined foreign language; speech synthesis means for converting the translated text into audio data; playback means for playing the audio data to the user; means for transmitting the text data obtained from the speech recognition means to a translation engine; means for transmitting the translated text obtained from the translation engine to a speech synthesis engine; and means for transmitting the audio data obtained from the speech synthesis engine to the user's terminal. This enables users who speak different languages ​​to communicate smoothly in real time.

[1214] A "user" refers to a person who uses a system to communicate.

[1215] "Means of voice input" refers to devices and technologies for converting a user's voice into a digital format.

[1216] "Digital format" refers to a state in which audio or data has been converted into a format that can be processed by electronic devices.

[1217] "Speech recognition means" refers to a function or technology that converts speech, which has been converted into a digital format, into text.

[1218] "Translation means" refers to a function or technology for translating text data into a specified foreign language.

[1219] "Speech synthesis means" refers to a function or technology for converting text data translated into a foreign language into speech data.

[1220] "Playback means" refers to devices and technologies for providing audio data to users as auditory information.

[1221] A "cloud-based speech recognition engine" refers to a speech recognition service provided via the internet.

[1222] "User's personal information terminal" refers to portable electronic devices such as smartphones and tablets.

[1223] "Audio output device" refers to a device that includes speakers or headphones for playing back audio data.

[1224] "Text data" refers to standard string information converted by speech recognition.

[1225] "Translation" refers to the text data after it has been translated.

[1226] "Audio data" refers to the format of digital audio generated by speech synthesis technology.

[1227] This invention is a system aimed at enabling users who speak different languages ​​to communicate smoothly in real time. The system includes a series of processes that convert audio into a digital format, convert it to text, translate it, convert it back into audio data, and finally play it back to the user.

[1228] System Configuration

[1229] This system consists of the following main elements:

[1230] 1. Voice input method

[1231] 2. Speech recognition means

[1232] 3. Translation methods

[1233] 4. Speech synthesis means

[1234] 5. Regeneration means

[1235] Hardware and software to use

[1236] For voice input, the user's mobile device, such as a smartphone or tablet, is used. These devices have built-in microphones that record the user's voice.

[1237] The speech recognition method sends recorded audio to a cloud-based speech recognition engine, which then converts the audio to text. Specifically, it uses APIs such as the Google Cloud Speech-to-Text API.

[1238] The translation method translates the text obtained by the speech recognition method into a specified foreign language. This uses a translation engine such as the Google Cloud Translation API.

[1239] The speech synthesis method converts translated text into speech data. A specific example used is the Microsoft Azure Text-to-Speech API.

[1240] The playback method involves playing the generated audio data to the user. The speaker or earphones built into the user's smartphone or tablet are used as the playback device.

[1241] Specific example

[1242] For example, the process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[1243] 1. The user says to their smartphone, "Hello, what's your name?"

[1244] 2. The device records the audio, converts it to a digital format, and saves it.

[1245] 3. The device sends the digital voice data to a cloud-based speech recognition engine (Google Cloud Speech-to-Text API).

[1246] 4. The server converts the audio data into text data.

[1247] 5. The server sends the text data to the translation engine (Google Cloud Translation API) for translation into English.

[1248] 6. The server sends the English text data to the text-to-speech engine (Microsoft Azure Text-to-Speech API) and converts it into speech data.

[1249] 7. The server sends the generated English audio data to the user's device.

[1250] 8. The device plays audio data, and an English voice says, "Hello, what is your name?"

[1251] Example of a prompt

[1252] The following are specific examples of prompt statements used in this system.

[1253] Prompt: Translate the Japanese phrase "Hello, what is your name?" into English, and then generate the English audio.

[1254] Using this prompt, the system goes through the processes of speech recognition, text translation, and speech synthesis to generate and provide results to the user.

[1255] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1256] Step 1:

[1257] Voice input

[1258] The user speaks into their smartphone or tablet.

[1259] Input: User's voice (e.g., "Hello, what's your name?")

[1260] Specific action: The user speaks into the microphone.

[1261] Output: Raw audio data

[1262] Step 2:

[1263] Audio recording and digital conversion

[1264] The device records the user's voice and converts it to a digital format.

[1265] Input: Raw audio data

[1266] Specific operation: The smartphone's audio engine (e.g., AudioRecord API) converts the raw audio into PCM digital data.

[1267] Output: Digital audio data in PCM format

[1268] Step 3:

[1269] Sending audio data

[1270] The device sends the converted digital audio data to a cloud-based speech recognition engine.

[1271] Input: PCM format digital audio data

[1272] Specific operation: Send digital audio data as an HTTP request to the Google Cloud Speech-to-Text API.

[1273] Output: Audio data sent to the server

[1274] Step 4:

[1275] Speech recognition

[1276] The server uses a speech recognition engine to convert digital audio data into text data.

[1277] Input: PCM format digital audio data

[1278] Specific operation: The Google Cloud Speech-to-Text API analyzes the audio data and converts it into text, "Hello, what is your name?".

[1279] Output: Text data "Hello, what is your name?"

[1280] Step 5:

[1281] Text translation

[1282] The server sends the text data acquired through speech recognition to the translation engine.

[1283] Input: Text data "Hello, what is your name?"

[1284] Specific operation: The server sends a translation request to the Google Cloud Translation API. The request includes the source text and the target language.

[1285] Output: English translation text data "Hello, what is your name?"

[1286] Step 6:

[1287] Preparation for speech synthesis

[1288] The server sends the translated English text to the speech synthesis engine.

[1289] Input: English text data "Hello, what is your name?"

[1290] Specific operation: The server sends a text-to-speech request to the Microsoft Azure Text-to-Speech API. The request includes the translated text and voice configuration information.

[1291] Output: Voice data request

[1292] Step 7:

[1293] Speech synthesis

[1294] The server uses a speech synthesis engine to convert text data into speech data.

[1295] Input: English text data "Hello, what is your name?"

[1296] Specific operation: The Microsoft Azure Text-to-Speech API parses English text and converts it into speech data.

[1297] Output: English audio data

[1298] Step 8:

[1299] Sending audio data

[1300] The server sends the generated English audio data to the user's device.

[1301] Input: English audio data

[1302] Specific operation: The server sends the generated audio data to the terminal as an HTTP response.

[1303] Output: Audio data sent to the user's terminal

[1304] Step 9:

[1305] Playback of audio data

[1306] The device plays the received audio data.

[1307] Input: English audio data

[1308] Specific action: The device's media player (e.g., MediaPlayer API) plays audio data, and the voice "Hello, what is your name?" is played through the speaker.

[1309] Output: Audio played to the user

[1310] (Application Example 1)

[1311] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1312] In brick-and-mortar retail environments, where smooth communication between customers and employees with different language backgrounds is essential, language barriers can prevent customers from obtaining necessary information, leading to a decline in service quality. Furthermore, the difficulty for employees to be fluent in multiple languages ​​reduces operational efficiency in physical stores. A system is needed to address these challenges and support communication in brick-and-mortar retail settings.

[1313] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1314] In this invention, the server includes means for inputting a user's voice and converting the voice into a digital format; voice recognition means for converting the voice converted into digital format into text; translation means for translating the text into a predetermined foreign language; voice synthesis means for converting the text translated into the foreign language into audio data; playback means for playing the audio data back to the user; means for supporting communication between customers and employees in a store; and means for customers or employees to speak into the device. This enables real-time voice translation and smooth communication between customers and employees who speak different languages.

[1315] "User" refers to the customers and employees who use this system.

[1316] "Means for inputting audio and converting said audio into a digital format" refers to a device or software that has the function of converting analog audio data into digital data.

[1317] "Speech recognition means" refers to a device or software that has the function of converting speech converted into digital format into text format.

[1318] "Translation means" refers to a device or software that has the function of translating text, which has been converted into text format by speech recognition means, into a predetermined foreign language.

[1319] "Speech synthesis means" refers to a device or software that has the function of converting translated text into speech data.

[1320] "Playback means" refers to a device or software that has the function of playing back audio data generated by a speech synthesis means to the user.

[1321] "Means for supporting communication between customers and employees within a store" refers to devices or software that provide the necessary functions for smooth communication between customers and employees who speak different languages ​​in a physical store.

[1322] "Device" refers to a personal information terminal (such as a smartphone or tablet) used for inputting and playing audio.

[1323] This invention is a multilingual real-time interpretation assistant system aimed at facilitating smooth communication between customers and employees who speak different languages. This invention is primarily implemented using mobile devices such as smartphones and tablets, and utilizes a cloud-based speech recognition engine and translation engine.

[1324] System Configuration

[1325] This system includes the following key hardware and software components:

[1326] 1. Mobile information terminal

[1327] It has a microphone and processor for voice input, recording, and conversion to digital format.

[1328] 2. Cloud-based speech recognition engine

[1329] Use the Google Cloud Speech-to-Text API to convert digital audio data into text data.

[1330] 3. Translation engine

[1331] The retrieved text data is translated into a specified foreign language using the Google Cloud Translate API.

[1332] 4. Speech synthesis engine

[1333] The Google Cloud Text-to-Speech API is used to convert translated text data into audio data.

[1334] 5. Regeneration means

[1335] Audio data is played using the speaker installed in the mobile device.

[1336] Processing flow details

[1337] The server includes means for inputting user voice and converting said voice into a digital format, voice recognition means, translation means, speech synthesis means, and playback means. By coordinating these means, real-time communication between customers and employees in a physical store is realized.

[1338] 1. Voice input and conversion

[1339] Personal digital assistant (PDA): The user (customer or employee) speaks into the device. The device records the audio and converts it to a digital format.

[1340] 2. Speech Recognition

[1341] Mobile device: Sends digital audio data to the Google Cloud Speech-to-Text API. The Google Cloud Speech-to-Text API converts the audio data into text data.

[1342] 3. Text Translation

[1343] Server: Sends text data to the Google Cloud Translate API for translation into the specified foreign language.

[1344] 4. Speech synthesis

[1345] Server: Sends the translated text data to the Google Cloud Text-to-Speech API and converts it into speech data.

[1346] 5. Audio Playback

[1347] Personal digital assistant (PDCA): Receives generated audio data and plays it through the speaker.

[1348] Specific example

[1349] For example, consider a case where a Japanese-speaking customer asks an English-speaking employee for the location of a product.

[1350] 1. Customer: "Where can I find this product?" they say to their smartphone.

[1351] 2. Personal digital assistant (PDCA): Records audio, converts it to a digital format, and sends it to the cloud.

[1352] 3. Google Cloud Speech-to-Text API: Converts audio data into text data such as "Where can I find this product?".

[1353] 4. Google Cloud Translate API: Translate the text data into "Where is this product located?".

[1354] 5. Google Cloud Text-to-Speech API: Converts translated text data into speech data.

[1355] 6. Mobile device: Play the English audio "Where is this product located?".

[1356] Example of a prompt

[1357] "Where can I find this product?"

[1358] In this way, customers and employees can communicate smoothly, overcoming language barriers.

[1359] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1360] Step 1:

[1361] Voice input and conversion to digital format

[1362] User: The user (customer or employee) speaks into a mobile device.

[1363] Input: Analog audio data.

[1364] Terminal: The microphone of the personal digital assistant (PDCA) records audio, and the analog audio data is converted into a digital format.

[1365] Output: Digital audio data.

[1366] Step 2:

[1367] Speech recognition of digital audio data

[1368] Terminal: Sends digital audio data to the cloud.

[1369] Input: Digital audio data.

[1370] Server (Google Cloud Speech-to-Text API): Receives digital audio data and performs speech recognition. Specifically, it analyzes audio patterns and converts them into text data.

[1371] Output: Text data (e.g., "Where can I find this product?").

[1372] Step 3:

[1373] Text data translation

[1374] Server: The translation engine (Google Cloud Translate API) translates text data into the specified foreign language.

[1375] Input: Japanese text data (e.g., "Where can I find this product?").

[1376] Server (Google Cloud Translate API): Receives text data and performs translation processing. Based on the model, it translates Japanese into the target language, such as English.

[1377] Output: Translated text data (e.g., "Where is this product located?").

[1378] Step 4:

[1379] Converting translated text to audio data

[1380] Server: Sends the translated text data to the text-to-speech engine (Google Cloud Text-to-Speech API) and converts it into speech data.

[1381] Input: Translated text data (e.g., "Where is this product located?").

[1382] Server (Google Cloud Text-to-Speech API): Converts text data into speech data. Specifically, it analyzes the text, breaks it down into phonemes, and then synthesizes them together.

[1383] Output: Audio data (e.g., "Where is this product located?").

[1384] Step 5:

[1385] Playback of audio data

[1386] Terminal: Receives audio data and plays it back to the user.

[1387] Input: Audio data received from the server (e.g., "Where is this product located?").

[1388] Device (speaker): Plays audio data.

[1389] Output: Voice response message (e.g., "Where is this product located?").

[1390] This processing step enables users to interact in real time with people who speak different languages.

[1391] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1392] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. It also provides a system that combines this with an emotion engine that recognizes the user's emotions and appropriately adjusts the translated expression accordingly. The invention will describe the system's program processing in natural language, and will also include specific usage examples.

[1393] System Overview

[1394] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system includes the following main elements:

[1395] 1. Voice input method

[1396] 2. Speech recognition means

[1397] 3. Translation methods

[1398] 4. Emotional Engine

[1399] 5. Speech synthesis means

[1400] 6. Reproduction means

[1401] System processing flow

[1402] 1. Voice input

[1403] User (terminal): The user speaks towards a smartphone or tablet. For example, the user makes a statement such as "Hello, what is your name?"

[1404] 2. Voice recognition

[1405] Terminal: Convert the voice input into digital format and save it locally.

[1406] Terminal: Transmit the saved digital voice data to a cloud-based voice recognition API.

[1407] Server: The voice recognition engine converts the digital voice data into text data "Hello, what is your name?"

[1408] 3. Emotion recognition

[1409] Server: The emotion engine analyzes the emotion from the user's voice in real time and generates emotion data. For example, it is recognized that the statement "Hello, what is your name?" is in a happy tone.

[1410] 4. Text translation

[1411] Server: Transmit the voice-recognized text data and emotion data to the translation engine.

[1412] Server: The translation engine translates the text data "Hello, what is your name?" into English as "Hello, what is your name?" and adjusts the expression to match the user's emotion.

[1413] 5. Speech synthesis

[1414] Server: Sends the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[1415] Server: The speech synthesis engine converts English text data into speech data.

[1416] Server: Sends the generated audio data to the terminal.

[1417] 6. Audio Playback

[1418] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play a cheerful voice saying, "Hello, what is your name?"

[1419] Specific example

[1420] For example, the processing flow when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[1421] 1. User: "Hello, what's your name?" they say to their smartphone.

[1422] 2. Device: Record audio, convert it to a digital format, and save it.

[1423] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[1424] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[1425] 5. Server: The emotion engine recognizes emotions from the user's voice and generates "happy" emotion data.

[1426] 6. Server: The translation engine translates the text into English as "Hello, what is your name?", maintaining the cheerful tone.

[1427] 7. Server: The speech synthesis engine converts English text into speech data.

[1428] 8. Server: Sends audio data to the user's terminal.

[1429] 9. Device: Play the voice data "Hello, what is your name?" in a cheerful tone.

[1430] 10. The other person responds, "My name is John."

[1431] 11. The other party's device: Record the audio, convert it to a digital format, and save it.

[1432] 12. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[1433] 13. Server: The speech recognition engine converts "My name is John." into text.

[1434] 14. Server: The emotion engine recognizes emotions from the other party's voice and generates "calm" emotion data.

[1435] 15. Server: The translation engine translates the text into Japanese as "My name is John," maintaining a calm tone.

[1436] 16. Server: The speech synthesis engine converts Japanese text into speech data.

[1437] 17. Server: Sends audio data to the user's terminal.

[1438] 18. Device: Play the audio data "My name is John." in a calm tone.

[1439] In this way, smooth and considerate communication between the user and the person speaking the foreign language is achieved.

[1440] The following describes the processing flow.

[1441] Step 1:

[1442] User (device): The user speaks to their smartphone or tablet and says, "Hello, what's your name?"

[1443] Step 2:

[1444] Terminal: The voice input application records the user's speech and converts it into a digital format.

[1445] Step 3:

[1446] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[1447] Step 4:

[1448] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[1449] Step 5:

[1450] Server: Sends text data to the emotion engine to analyze the user's emotions.

[1451] Step 6:

[1452] Server: The emotion engine analyzes the tone and pitch of the voice, etc., and recognizes that the user has a "happy" emotion. It also generates emotion data.

[1453] Step 7:

[1454] Server: Transmits the voice-recognized text data and emotion data to the translation engine.

[1455] Step 8:

[1456] Server: The translation engine translates the text data "Hello, what is your name?" into English as "Hello, what is your name?" and adjusts the expression to fit the user's "happy" emotion.

[1457] Step 9:

[1458] Server: Transmits the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[1459] Step 10:

[1460] Server: The speech synthesis engine converts the English text data into speech data.

[1461] Step 11:

[1462] Server: Transmits the generated speech data to the terminal.

[1463] Step 12:

[1464] Terminal: Receives the speech data and plays it as speech to the user. For example, the speech is played in a happy tone like "Hello, what is your name?".

[1465] Step 13:

[1466] The other party (device): The other party responds with "My name is John."

[1467] Step 14:

[1468] Recipient's device: Records the recipient's speech and converts it to a digital format.

[1469] Step 15:

[1470] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[1471] Step 16:

[1472] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[1473] Step 17:

[1474] Server: Sends text data to the emotion engine to analyze the other party's emotions.

[1475] Step 18:

[1476] Server: The emotion engine analyzes the tone and pitch of the voice and recognizes that the other party has a "calm" emotion. It also generates emotion data.

[1477] Step 19:

[1478] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[1479] Step 20:

[1480] Server: The translation engine translates the text data "My name is John." into Japanese as "私の名前はジョンです。" and adjusts the expression to fit the "calm" emotion of the recipient.

[1481] Step 21:

[1482] Server: Send the translated Japanese text data "私の名前はジョンです。" to the speech synthesis engine.

[1483] Step 22:

[1484] Server: The speech synthesis engine converts the Japanese text data into speech data.

[1485] Step 23:

[1486] Server: Send the generated speech data to the user's terminal.

[1487] Step 24:

[1488] Terminal: Receive the speech data and play it as speech to the user. For example, the speech is played in a calm tone as "私の名前はジョンです。".

[1489] In this way, smooth communication that takes into account emotions is realized between the user and the foreign language speaker.

[1490] (Example 2)

[1491] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset type terminal 314 is referred to as the "terminal".

[1492] In recent years, international communication has increased, and smooth multilingual conversation is required. However, conventional translation systems simply translate languages ​​without adjusting the translation to take into account the speaker's emotions, resulting in a decline in the quality of communication. Furthermore, in real-time speech translation, the processing speed from speech recognition to translation and playback has been a problem. Therefore, there is a need for a system that accurately reflects the speaker's emotions while enabling smooth, real-time communication between multiple languages.

[1493] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1494] In this invention, the server includes means for recognizing emotions from the user's voice in real time and generating emotion data, means for adjusting translated expressions based on translated text and emotion data, and means for converting text translated into a foreign language into speech data. This enables real-time multilingual voice communication using translated expressions that reflect the speaker's emotions.

[1495] A "user" refers to a person who uses the system to perform voice input and voice translation.

[1496] "Digital format" refers to a format in which analog audio is converted into digital data that can be processed by a computer.

[1497] "Speech recognition means" refers to means that have the function of converting digital speech data into text data.

[1498] "Translation means" refers to a means that has the function of converting text from one language to another language.

[1499] "Emotion recognition means" refers to means that have the function of analyzing the speaker's emotions in real time from audio data and generating emotion data.

[1500] "Means for adjusting translated expressions" refers to means that have the function of adjusting translated expressions to reflect appropriate sentiment based on the translated text and sentiment data.

[1501] "Speech synthesis means" refers to means that have the function of converting text data into speech data.

[1502] "Playback means" refers to means that have the function of outputting the generated audio data as sound.

[1503] A "cloud-based speech recognition engine" refers to an engine that uses servers and software accessible via a network to convert speech data into text data.

[1504] "Communication equipment" refers to electronic devices with communication capabilities, such as smartphones and tablets.

[1505] "Audio equipment" refers to devices such as speakers and earphones used to play audio data.

[1506] This invention is a speech translation system that enables users to communicate smoothly with people who speak a foreign language. The system consists of the following processes: speech input, speech recognition, emotion recognition, text translation, speech synthesis, and speech playback. Each process utilizes specific hardware and software.

[1507] System Configuration

[1508] 1. Voice input

[1509] User: Speaks into a smartphone or tablet. For example, says, "Hello, what's your name?"

[1510] Terminal: Uses a microphone as an interface to convert the user's voice into a digital format.

[1511] 2. Speech Recognition

[1512] Terminal: Sends digital voice data to a cloud-based speech recognition engine. The API used is the speech recognition engine.

[1513] Server: Converts the digital audio data received by the speech recognition engine into text data.

[1514] 3. Emotion recognition

[1515] Server: The emotion recognition engine analyzes the user's voice in real time and generates emotion data. The emotion recognition engine used is an emotion analysis engine.

[1516] 4. Text Translation

[1517] Server: Sends the speech-recognized text data and generated sentiment data to the translation engine. The translation engine used is a text translation engine.

[1518] Server: The translation engine converts the text into the target language and adjusts the translation based on sentiment data.

[1519] 5. Speech synthesis

[1520] Server: Sends the translated text data to the speech synthesis engine. The speech synthesis engine used is speech synthesis software.

[1521] Server: The speech synthesis engine converts text data into speech data.

[1522] 6. Audio Playback

[1523] Terminal: Receives audio data and plays it back using the terminal's sound device. For example, it might play back "Hello, what is your name?" in a cheerful tone.

[1524] Specific example

[1525] As a concrete example, we will simulate a conversation between a Japanese-speaking user and an English-speaking partner. The following is the processing flow.

[1526] 1. User: "Hello, what's your name?" they say to their smartphone.

[1527] 2. Device: Records audio and converts it into digital data.

[1528] 3. Server: The speech recognition engine converts "Hello, what is your name?" into text data.

[1529] 4. Server: The emotion recognition engine generates text data and emotion data, and detects the emotion of "happiness".

[1530] 5. Server: The translation engine translates the text data into English as "Hello, what is your name?", maintaining a cheerful tone.

[1531] 6. Server: The speech synthesis engine converts the translated text into speech data.

[1532] 7. Terminal: Receives audio data and plays "Hello, what is your name?" in a cheerful tone.

[1533] Example of a prompt

[1534] Use the following prompts to input data into the generative AI model.

[1535] "Please translate the Japanese phrase 'Hello, what is your name?' into English and convert it into speech, while maintaining the emotion in the speech."

[1536] "Translate the English sentence 'My name is John.' into Japanese and then speak it in a calm tone."

[1537] In this way, real-time voice translation that takes emotions into consideration is achieved between the user and the person speaking the foreign language.

[1538] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1539] Program processing flow

[1540] Step 1: Voice Input

[1541] User: Speak into the microphone on your smartphone or tablet and say, "Hello, what's your name?"

[1542] Input: Analog audio

[1543] Output: Digital audio data

[1544] Terminal: Captures the user's voice with a microphone, converts the analog audio to a digital format using an audio digitization module, and saves it to a buffer.

[1545] Step 2: Speech Recognition

[1546] Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[1547] Input: Digital audio data

[1548] Output: Text data ("Hello, what is your name?")

[1549] Server: The speech recognition engine analyzes the digital speech data it receives and converts it into text data, taking into account phonemes and context.

[1550] Step 3: Emotion Recognition

[1551] Server: Sends the speech-recognized text data and digital audio data to the emotion recognition engine.

[1552] Input: Text data, digital audio data

[1553] Output: Emotional data (e.g., "Happy")

[1554] Server: The emotion recognition engine analyzes the tone, pitch, and speed of the voice, evaluates the user's emotional state in real time, and generates emotion data.

[1555] Step 4: Text Translation

[1556] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[1557] Input: Text data, sentiment data

[1558] Output: Translated text data (e.g., "Hello, what is your name?")

[1559] Server: The translation engine converts text data into the target foreign language and adjusts the translation based on sentiment data.

[1560] Step 5: Speech Synthesis

[1561] Server: Sends the translated text data to the speech synthesis engine.

[1562] Input: Translated text data

[1563] Output: Audio data

[1564] Server: The speech synthesis engine analyzes the translated text data and generates speech data that reflects emotional tone.

[1565] Step 6: Audio Playback

[1566] Device: Receives audio data and plays it through the device's speaker.

[1567] Input: Audio data

[1568] Output: Played audio (e.g., "Hello, what is your name?")

[1569] Terminal: Uses an audio device to play the generated audio data to the user.

[1570] (Application Example 2)

[1571] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1572] In physical stores, communication between customers who speak different languages ​​and employees often breaks down. This can lead to decreased customer satisfaction and a decline in the quality of service. Furthermore, it is difficult for employees to immediately understand and respond to multiple languages, especially when it comes to conveying nuances, including emotions. In this situation, it is necessary to improve customer service in physical stores by streamlining foreign language support and providing accurate translations that take emotions into consideration.

[1573] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1574] This invention includes a server that includes means for inputting a user's voice and converting the voice into a digital format, speech recognition means for converting the digitally converted voice into text, translation means for translating the text into a predetermined foreign language, an emotion engine that recognizes the emotion of the translated text and adjusts the translated expression appropriately according to the emotion, speech synthesis means for converting the translated text into audio data, playback means for playing the audio data back to the user, and a device that implements an application that supports real-time multilingual communication between the user and customers who speak a foreign language. This makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the emotions of customers.

[1575] "Means of inputting audio and converting it to a digital format" refers to devices or processes that receive audio signals and record them as digital data.

[1576] "Speech recognition means" refers to a technology or device that analyzes speech converted into digital data and converts it into corresponding text data.

[1577] "Translation methods" refer to software or algorithms used to convert text data from one language into another language.

[1578] An "emotion engine" is a technology or algorithm that recognizes a speaker's emotions from audio or text data and analyzes that emotional information.

[1579] "Speech synthesis means" refers to a technology or device that analyzes text data and generates speech data corresponding to that text.

[1580] "Reproduction means" refers to a device or technology used to actually pronounce the generated audio data as sound.

[1581] "Device" is a general term for any equipment on which an application is installed, such as communication terminals, smartphones, and smart glasses.

[1582] A "cloud-based speech recognition engine" refers to a service or technology for performing speech recognition in a cloud computing environment.

[1583] A "communication terminal" refers to an electronic device such as a smartphone or tablet that has functions such as voice input and playback, and data communication.

[1584] An "application that supports multilingual communication" is software designed to enable users who speak different languages ​​to communicate with each other in real time.

[1585] This invention is a system that facilitates multilingual support in physical stores and provides appropriate translations that take emotions into consideration. This system uses a device to support real-time communication between users and foreign language-speaking customers. The main elements of the system are means for inputting voice and converting it to a digital format, voice recognition means, translation means, emotion engine, voice synthesis means, playback means, and a device for implementing the application.

[1586] The following hardware and software will be used to implement this system:

[1587] Hardware: Communication devices (e.g., smartphones, tablets, smart glasses)

[1588] software:

[1589] Speech recognition: Google Cloud Speech-to-Text API

[1590] Translation: Google Cloud Translation API

[1591] Emotion recognition: IBM Watson Tone Analyzer API

[1592] Text-to-speech: Google Cloud Text-to-Speech API

[1593] Real-time communication: Firebase Realtime Database

[1594] Description of the system's program processing

[1595] First, the user speaks into a communication device. The communication device converts this audio into a digital format and uploads it to Firebase. The server sends the audio data stored in Firebase to the Google Cloud Speech-to-Text API, where it converts the audio into text data.

[1596] Next, the server sends this text data to the IBM Watson Tone Analyzer API for sentiment analysis. The text, including sentiment data, is then sent to the Google Cloud Translation API for translation into the specified foreign language. Based on the sentiment data, the translated sentence is adjusted to an appropriate tone.

[1597] The server then sends the translated text data to the Google Cloud Text-to-Speech API to generate audio data. This audio data is then stored again in Firebase and sent to the user's communication device. The communication device plays this audio data, allowing the user to provide the translated message to their customers.

[1598] Specific example

[1599] Let's consider a situation where an employee working in a souvenir shop in a tourist area needs to communicate smoothly with foreign customers. For example,

[1600] User (employee): "Welcome, how can I help you?" they say to their smartphone.

[1601] System: The server recognizes the speech, analyzes the emotions, translates it into English, and plays back "Welcome! How can I help you?" in a friendly tone.

[1602] The other party (customer) asks, "Can you tell me where I can find local souvenirs?"

[1603] User: Provide a properly translated response through the system again to facilitate smooth communication.

[1604] Example of a prompt

[1605] "Based on the following information, please generate code that performs translations tailored to the customer's emotions."

[1606] Customer comment: "Hello, can I purchase this?"

[1607] Emotion: "Friendly"

[1608] Translated statement: "Hello, can I purchase this?"

[1609] Code for speech synthesis in a friendly tone

[1610] This invention makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the customer's feelings.

[1611] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1612] Step 1:

[1613] Input: The user speaks into the communication terminal.

[1614] Operation: The user speaks to the communication terminal saying, "Welcome, how can I help you?"

[1615] Data processing: The communication terminal converts this audio into a digital format.

[1616] Output: Digital audio data is generated.

[1617] Step 2:

[1618] Input: Digital audio data.

[1619] Operation: The communication device uploads this digital audio data to Firebase.

[1620] Data processing: Digital audio data is transferred to the cloud.

[1621] Output: Digital audio data is stored on Firebase.

[1622] Step 3:

[1623] Input: Digital audio data stored in Firebase.

[1624] Operation: The server retrieves audio data from Firebase and sends it to the Google Cloud Speech-to-Text API.

[1625] Data processing: The speech recognition engine analyzes the digital audio data and converts it into text data.

[1626] Output: The text data "Welcome, how can I help you?" is generated.

[1627] Step 4:

[1628] Input: Text data.

[1629] Operation: The server sends text data to the IBM Watson Tone Analyzer API.

[1630] Data processing: The emotion engine analyzes the text data and generates emotion data (e.g., "friendly tone").

[1631] Output: Sentiment data is generated and attached to the text data.

[1632] Step 5:

[1633] Input: Text data and sentiment data.

[1634] Operation: The server sends this data to the Google Cloud Translation API.

[1635] Data processing: The translation engine translates text data into the specified foreign language and adjusts it based on sentiment data.

[1636] Output: The translated text data "Welcome! How can I help you?" is generated.

[1637] Step 6:

[1638] Input: Translated text data.

[1639] Operation: The server sends this text data to the Google Cloud Text-to-Speech API.

[1640] Data processing: The speech synthesis engine converts the translated text data into speech data.

[1641] Output: The audio data "Welcome! How can I help you?" is generated.

[1642] Step 7:

[1643] Input: Translated audio data.

[1644] Operation: The server uploads the generated audio data to Firebase and sends it to the user's communication device.

[1645] Data processing: Audio data is transferred from the server to the communication terminal.

[1646] Output: The audio data is saved to the communication terminal.

[1647] Step 8:

[1648] Input: Translated audio data.

[1649] Operation: The communication terminal plays the translated audio data through its speaker.

[1650] Data processing: Audio data is output as audio.

[1651] Output: The translated audio "Welcome! How can I help you?" is played to the customer.

[1652] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1653] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1654] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1655] [Fourth Embodiment]

[1656] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1657] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1658] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1659] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1660] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1661] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1662] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1663] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1664] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1665] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1666] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1667] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1668] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1669] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. The system's program processing will be described in natural language to illustrate embodiments of the invention. Specific usage examples will also be provided.

[1670] System Overview

[1671] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system consists of the following main elements:

[1672] 1. Voice input method

[1673] 2. Speech recognition means

[1674] 3. Translation methods

[1675] 4. Speech synthesis means

[1676] 5. Regeneration means

[1677] System processing flow

[1678] 1. Voice input

[1679] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[1680] 2. Speech Recognition

[1681] Terminal: Converts voice input to a digital format and saves it locally.

[1682] Terminal: Sends stored digital audio data to a cloud-based speech recognition engine.

[1683] Server: The speech recognition engine converts the digital speech data into text data "Hello, what is your name?".

[1684] 3. Text Translation

[1685] Server: Sends the speech-recognized text to the translation engine.

[1686] Server: The translation engine translates the text into English as "Hello, what is your name?".

[1687] 4. Speech synthesis

[1688] Server: Sends the translated English text "Hello, what is your name?" to the speech synthesis engine.

[1689] Server: The speech synthesis engine converts English text into speech data.

[1690] Server: Sends the generated audio data to the terminal.

[1691] 5. Audio Playback

[1692] Terminal: Receives audio data and plays it back to the user.

[1693] Specific example

[1694] For example, the specific process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[1695] 1. User: "Hello, what's your name?" they say to their smartphone.

[1696] 2. Device: Record audio, convert it to a digital format, and save it.

[1697] 3. Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[1698] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[1699] 5. Server: The translation engine translates the text into English as "Hello, what is your name?".

[1700] 6. Server: The speech synthesis engine converts English text into speech data.

[1701] 7. Server: Sends audio data to the user's terminal.

[1702] 8. Terminal: Play the audio data "Hello, what is your name?" to the user.

[1703] 9. The other person responds, "My name is John."

[1704] 10. The other party's device: Record the audio, convert it to a digital format, and save it.

[1705] 11. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[1706] 12. Server: The speech recognition engine converts "My name is John." into text.

[1707] 13. Server: The translation engine translates the text into Japanese as "My name is John."

[1708] 14. Server: The speech synthesis engine converts Japanese text into speech data.

[1709] 15. Server: Sends audio data to the user's terminal.

[1710] 16. Terminal: Play the audio data "My name is John." to the user.

[1711] The above process enables smooth communication between users and people who speak a foreign language.

[1712] The following describes the processing flow.

[1713] Step 1:

[1714] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[1715] Step 2:

[1716] Terminal: The voice input application records the user's speech and converts it into a digital format.

[1717] Step 3:

[1718] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[1719] Step 4:

[1720] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[1721] Step 5:

[1722] Server: Sends the speech-recognized text data to the translation engine.

[1723] Step 6:

[1724] Server: The translation engine translates the text data into English as "Hello, what is your name?".

[1725] Step 7:

[1726] Server: Sends the translated English text data to the speech synthesis engine.

[1727] Step 8:

[1728] Server: The speech synthesis engine converts English text data into speech data.

[1729] Step 9:

[1730] Server: Sends the generated audio data to the terminal.

[1731] Step 10:

[1732] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio "Hello, what is your name?".

[1733] Step 11:

[1734] The other party (device): The other party responds with "My name is John."

[1735] Step 12:

[1736] Recipient's device: Records the recipient's speech and converts it to a digital format.

[1737] Step 13:

[1738] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[1739] Step 14:

[1740] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[1741] Step 15:

[1742] Server: Sends the speech-recognized text data to the translation engine.

[1743] Step 16:

[1744] Server: The translation engine translates the text data into Japanese as "My name is John."

[1745] Step 17:

[1746] Server: Sends the translated Japanese text data to the speech synthesis engine.

[1747] Step 18:

[1748] Server: The speech synthesis engine converts Japanese text data into speech data.

[1749] Step 19:

[1750] Server: Sends the generated audio data to the user's device.

[1751] Step 20:

[1752] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play the audio saying, "My name is John."

[1753] This series of steps enables real-time voice translation between the user and the person speaking the foreign language, facilitating smooth communication.

[1754] (Example 1)

[1755] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1756] In modern society, smooth communication between people who speak different languages ​​is a challenging task. In particular, there is a lack of technology for real-time translation and seamless multilingual conversation. Therefore, there is a need to develop systems that allow users who speak different languages ​​to communicate easily.

[1757] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1758] In this invention, the server includes means for inputting the user's voice and converting the voice into a digital format; speech recognition means for converting the digitally converted voice into text; translation means for translating the text into a predetermined foreign language; speech synthesis means for converting the translated text into audio data; playback means for playing the audio data to the user; means for transmitting the text data obtained from the speech recognition means to a translation engine; means for transmitting the translated text obtained from the translation engine to a speech synthesis engine; and means for transmitting the audio data obtained from the speech synthesis engine to the user's terminal. This enables users who speak different languages ​​to communicate smoothly in real time.

[1759] A "user" refers to a person who uses a system to communicate.

[1760] "Means of voice input" refers to devices and technologies for converting a user's voice into a digital format.

[1761] "Digital format" refers to a state in which audio or data has been converted into a format that can be processed by electronic devices.

[1762] "Speech recognition means" refers to a function or technology that converts speech, which has been converted into a digital format, into text.

[1763] "Translation means" refers to a function or technology for translating text data into a specified foreign language.

[1764] "Speech synthesis means" refers to a function or technology for converting text data translated into a foreign language into speech data.

[1765] "Playback means" refers to devices and technologies for providing audio data to users as auditory information.

[1766] A "cloud-based speech recognition engine" refers to a speech recognition service provided via the internet.

[1767] "User's personal information terminal" refers to portable electronic devices such as smartphones and tablets.

[1768] "Audio output device" refers to a device that includes speakers or headphones for playing back audio data.

[1769] "Text data" refers to standard string information converted by speech recognition.

[1770] "Translation" refers to the text data after it has been translated.

[1771] "Audio data" refers to the format of digital audio generated by speech synthesis technology.

[1772] This invention is a system aimed at enabling users who speak different languages ​​to communicate smoothly in real time. The system includes a series of processes that convert audio into a digital format, convert it to text, translate it, convert it back into audio data, and finally play it back to the user.

[1773] System Configuration

[1774] This system consists of the following main elements:

[1775] 1. Voice input method

[1776] 2. Speech recognition means

[1777] 3. Translation methods

[1778] 4. Speech synthesis means

[1779] 5. Regeneration means

[1780] Hardware and software to use

[1781] For voice input, the user's mobile device, such as a smartphone or tablet, is used. These devices have built-in microphones that record the user's voice.

[1782] The speech recognition method sends recorded audio to a cloud-based speech recognition engine, which then converts the audio to text. Specifically, it uses APIs such as the Google Cloud Speech-to-Text API.

[1783] The translation method translates the text obtained by the speech recognition method into a specified foreign language. This uses a translation engine such as the Google Cloud Translation API.

[1784] The speech synthesis method converts translated text into speech data. A specific example used is the Microsoft Azure Text-to-Speech API.

[1785] The playback method involves playing the generated audio data to the user. The speaker or earphones built into the user's smartphone or tablet are used as the playback device.

[1786] Specific example

[1787] For example, the process when an English-speaking person and a Japanese-speaking user have a conversation is as follows:

[1788] 1. The user says to their smartphone, "Hello, what's your name?"

[1789] 2. The device records the audio, converts it to a digital format, and saves it.

[1790] 3. The device sends the digital voice data to a cloud-based speech recognition engine (Google Cloud Speech-to-Text API).

[1791] 4. The server converts the audio data into text data.

[1792] 5. The server sends the text data to the translation engine (Google Cloud Translation API) for translation into English.

[1793] 6. The server sends the English text data to the text-to-speech engine (Microsoft Azure Text-to-Speech API) and converts it into speech data.

[1794] 7. The server sends the generated English audio data to the user's device.

[1795] 8. The device plays audio data, and an English voice says, "Hello, what is your name?"

[1796] Example of a prompt

[1797] The following are specific examples of prompt statements used in this system.

[1798] Prompt: Translate the Japanese phrase "Hello, what is your name?" into English, and then generate the English audio.

[1799] Using this prompt, the system goes through the processes of speech recognition, text translation, and speech synthesis to generate and provide results to the user.

[1800] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1801] Step 1:

[1802] Voice input

[1803] The user speaks into their smartphone or tablet.

[1804] Input: User's voice (e.g., "Hello, what's your name?")

[1805] Specific action: The user speaks into the microphone.

[1806] Output: Raw audio data

[1807] Step 2:

[1808] Audio recording and digital conversion

[1809] The device records the user's voice and converts it to a digital format.

[1810] Input: Raw audio data

[1811] Specific operation: The smartphone's audio engine (e.g., AudioRecord API) converts the raw audio into PCM digital data.

[1812] Output: Digital audio data in PCM format

[1813] Step 3:

[1814] Sending audio data

[1815] The device sends the converted digital audio data to a cloud-based speech recognition engine.

[1816] Input: PCM format digital audio data

[1817] Specific operation: Send digital audio data as an HTTP request to the Google Cloud Speech-to-Text API.

[1818] Output: Audio data sent to the server

[1819] Step 4:

[1820] Speech recognition

[1821] The server uses a speech recognition engine to convert digital audio data into text data.

[1822] Input: PCM format digital audio data

[1823] Specific operation: The Google Cloud Speech-to-Text API analyzes the audio data and converts it into text, "Hello, what is your name?".

[1824] Output: Text data "Hello, what is your name?"

[1825] Step 5:

[1826] Text translation

[1827] The server sends the text data acquired through speech recognition to the translation engine.

[1828] Input: Text data "Hello, what is your name?"

[1829] Specific operation: The server sends a translation request to the Google Cloud Translation API. The request includes the source text and the target language.

[1830] Output: English translation text data "Hello, what is your name?"

[1831] Step 6:

[1832] Preparation for speech synthesis

[1833] The server sends the translated English text to the speech synthesis engine.

[1834] Input: English text data "Hello, what is your name?"

[1835] Specific operation: The server sends a text-to-speech request to the Microsoft Azure Text-to-Speech API. The request includes the translated text and voice configuration information.

[1836] Output: Voice data request

[1837] Step 7:

[1838] Speech synthesis

[1839] The server uses a speech synthesis engine to convert text data into speech data.

[1840] Input: English text data "Hello, what is your name?"

[1841] Specific operation: The Microsoft Azure Text-to-Speech API parses English text and converts it into speech data.

[1842] Output: English audio data

[1843] Step 8:

[1844] Sending audio data

[1845] The server sends the generated English audio data to the user's device.

[1846] Input: English audio data

[1847] Specific operation: The server sends the generated audio data to the terminal as an HTTP response.

[1848] Output: Audio data sent to the user's terminal

[1849] Step 9:

[1850] Playback of audio data

[1851] The device plays the received audio data.

[1852] Input: English audio data

[1853] Specific action: The device's media player (e.g., MediaPlayer API) plays audio data, and the voice "Hello, what is your name?" is played through the speaker.

[1854] Output: Audio played to the user

[1855] (Application Example 1)

[1856] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1857] In brick-and-mortar retail environments, where smooth communication between customers and employees with different language backgrounds is essential, language barriers can prevent customers from obtaining necessary information, leading to a decline in service quality. Furthermore, the difficulty for employees to be fluent in multiple languages ​​reduces operational efficiency in physical stores. A system is needed to address these challenges and support communication in brick-and-mortar retail settings.

[1858] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1859] In this invention, the server includes means for inputting a user's voice and converting the voice into a digital format; voice recognition means for converting the voice converted into digital format into text; translation means for translating the text into a predetermined foreign language; voice synthesis means for converting the text translated into the foreign language into audio data; playback means for playing the audio data back to the user; means for supporting communication between customers and employees in a store; and means for customers or employees to speak into the device. This enables real-time voice translation and smooth communication between customers and employees who speak different languages.

[1860] "User" refers to the customers and employees who use this system.

[1861] "Means for inputting audio and converting said audio into a digital format" refers to a device or software that has the function of converting analog audio data into digital data.

[1862] "Speech recognition means" refers to a device or software that has the function of converting speech converted into digital format into text format.

[1863] "Translation means" refers to a device or software that has the function of translating text, which has been converted into text format by speech recognition means, into a predetermined foreign language.

[1864] "Speech synthesis means" refers to a device or software that has the function of converting translated text into speech data.

[1865] "Playback means" refers to a device or software that has the function of playing back audio data generated by a speech synthesis means to the user.

[1866] "Means for supporting communication between customers and employees within a store" refers to devices or software that provide the necessary functions for smooth communication between customers and employees who speak different languages ​​in a physical store.

[1867] "Device" refers to a personal information terminal (such as a smartphone or tablet) used for inputting and playing audio.

[1868] This invention is a multilingual real-time interpretation assistant system aimed at facilitating smooth communication between customers and employees who speak different languages. This invention is primarily implemented using mobile devices such as smartphones and tablets, and utilizes a cloud-based speech recognition engine and translation engine.

[1869] System Configuration

[1870] This system includes the following key hardware and software components:

[1871] 1. Mobile information terminal

[1872] It has a microphone and processor for voice input, recording, and conversion to digital format.

[1873] 2. Cloud-based speech recognition engine

[1874] Use the Google Cloud Speech-to-Text API to convert digital audio data into text data.

[1875] 3. Translation engine

[1876] The retrieved text data is translated into a specified foreign language using the Google Cloud Translate API.

[1877] 4. Speech synthesis engine

[1878] The Google Cloud Text-to-Speech API is used to convert translated text data into audio data.

[1879] 5. Regeneration means

[1880] Audio data is played using the speaker installed in the mobile device.

[1881] Processing flow details

[1882] The server includes means for inputting user voice and converting said voice into a digital format, voice recognition means, translation means, speech synthesis means, and playback means. By coordinating these means, real-time communication between customers and employees in a physical store is realized.

[1883] 1. Voice input and conversion

[1884] Personal digital assistant (PDA): The user (customer or employee) speaks into the device. The device records the audio and converts it to a digital format.

[1885] 2. Speech Recognition

[1886] Mobile device: Sends digital audio data to the Google Cloud Speech-to-Text API. The Google Cloud Speech-to-Text API converts the audio data into text data.

[1887] 3. Text Translation

[1888] Server: Sends text data to the Google Cloud Translate API for translation into the specified foreign language.

[1889] 4. Speech synthesis

[1890] Server: Sends the translated text data to the Google Cloud Text-to-Speech API and converts it into speech data.

[1891] 5. Audio Playback

[1892] Personal digital assistant (PDCA): Receives generated audio data and plays it through the speaker.

[1893] Specific example

[1894] For example, consider a case where a Japanese-speaking customer asks an English-speaking employee for the location of a product.

[1895] 1. Customer: "Where can I find this product?" they say to their smartphone.

[1896] 2. Personal digital assistant (PDCA): Records audio, converts it to a digital format, and sends it to the cloud.

[1897] 3. Google Cloud Speech-to-Text API: Converts audio data into text data such as "Where can I find this product?".

[1898] 4. Google Cloud Translate API: Translate the text data into "Where is this product located?".

[1899] 5. Google Cloud Text-to-Speech API: Converts translated text data into speech data.

[1900] 6. Mobile device: Play the English audio "Where is this product located?".

[1901] Example of a prompt

[1902] "Where can I find this product?"

[1903] In this way, customers and employees can communicate smoothly, overcoming language barriers.

[1904] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1905] Step 1:

[1906] Voice input and conversion to digital format

[1907] User: The user (customer or employee) speaks into a mobile device.

[1908] Input: Analog audio data.

[1909] Terminal: The microphone of the personal digital assistant (PDCA) records audio, and the analog audio data is converted into a digital format.

[1910] Output: Digital audio data.

[1911] Step 2:

[1912] Speech recognition of digital audio data

[1913] Terminal: Sends digital audio data to the cloud.

[1914] Input: Digital audio data.

[1915] Server (Google Cloud Speech-to-Text API): Receives digital audio data and performs speech recognition. Specifically, it analyzes audio patterns and converts them into text data.

[1916] Output: Text data (e.g., "Where can I find this product?").

[1917] Step 3:

[1918] Text data translation

[1919] Server: The translation engine (Google Cloud Translate API) translates text data into the specified foreign language.

[1920] Input: Japanese text data (e.g., "Where can I find this product?").

[1921] Server (Google Cloud Translate API): Receives text data and performs translation processing. Based on the model, it translates Japanese into the target language, such as English.

[1922] Output: Translated text data (e.g., "Where is this product located?").

[1923] Step 4:

[1924] Converting translated text to audio data

[1925] Server: Sends the translated text data to the text-to-speech engine (Google Cloud Text-to-Speech API) and converts it into speech data.

[1926] Input: Translated text data (e.g., "Where is this product located?").

[1927] Server (Google Cloud Text-to-Speech API): Converts text data into speech data. Specifically, it analyzes the text, breaks it down into phonemes, and then synthesizes them together.

[1928] Output: Audio data (e.g., "Where is this product located?").

[1929] Step 5:

[1930] Playback of audio data

[1931] Terminal: Receives audio data and plays it back to the user.

[1932] Input: Audio data received from the server (e.g., "Where is this product located?").

[1933] Device (speaker): Plays audio data.

[1934] Output: Voice response message (e.g., "Where is this product located?").

[1935] This processing step enables users to interact in real time with people who speak different languages.

[1936] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1937] This invention relates to a system that facilitates communication in foreign languages ​​and translates speech in real time. It also provides a system that combines this with an emotion engine that recognizes the user's emotions and appropriately adjusts the translated expression accordingly. The invention will describe the system's program processing in natural language, and will also include specific usage examples.

[1938] System Overview

[1939] The system of this invention is designed to enable users to communicate smoothly with people who speak a foreign language. This system includes the following main elements:

[1940] 1. Voice input method

[1941] 2. Speech recognition means

[1942] 3. Translation methods

[1943] 4. Emotional Engine

[1944] 5. Speech synthesis means

[1945] 6. Regeneration means

[1946] System processing flow

[1947] 1. Voice input

[1948] User (device): The user speaks into their smartphone or tablet. For example, they might say, "Hello, what's your name?"

[1949] 2. Speech Recognition

[1950] Terminal: Converts voice input to a digital format and saves it locally.

[1951] Terminal: Sends stored digital audio data to a cloud-based speech recognition API.

[1952] Server: The speech recognition engine converts the digital speech data into text data "Hello, what is your name?".

[1953] 3. Emotion recognition

[1954] Server: The emotion engine analyzes the user's voice in real time and generates emotion data. For example, it recognizes that the phrase "Hello, what's your name?" is said in a happy tone.

[1955] 4. Text Translation

[1956] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[1957] Server: The translation engine translates the text data "こんにちは、お名前は何ですか?" to English as "Hello, what is your name?" and adjusts the expression to fit the user's emotion.

[1958] 5. Voice Synthesis

[1959] Server: Send the translated English text data "Hello, what is your name?" to the voice synthesis engine.

[1960] Server: The voice synthesis engine converts the English text data into voice data.

[1961] Server: Send the generated voice data to the terminal.

[1962] 6. Voice Playback

[1963] Terminal: Receive the voice data and play it as voice for the user. For example, the voice "Hello, what is your name?" is played in a happy tone.

[1964] Specific Example

[1965] For example, the processing flow when a user who speaks Japanese converses with a person who speaks English is as follows.

[1966] 1. User: Speak towards the smartphone "こんにちは、お名前は何ですか?".

[1967] 2. Terminal: Record the voice, convert it to digital format and save it.

[1968] 3. Terminal: Send the digital voice data to a cloud-based voice recognition engine.

[1969] 4. Server: The speech recognition engine converts "Hello, what is your name?" into text.

[1970] 5. Server: The emotion engine recognizes emotions from the user's voice and generates "happy" emotion data.

[1971] 6. Server: The translation engine translates the text into English as "Hello, what is your name?", maintaining the cheerful tone.

[1972] 7. Server: The speech synthesis engine converts English text into speech data.

[1973] 8. Server: Sends audio data to the user's terminal.

[1974] 9. Device: Play the voice data "Hello, what is your name?" in a cheerful tone.

[1975] 10. The other person responds, "My name is John."

[1976] 11. The other party's device: Record the audio, convert it to a digital format, and save it.

[1977] 12. The recipient's device: Sends digital voice data to a cloud-based speech recognition engine.

[1978] 13. Server: The speech recognition engine converts "My name is John." into text.

[1979] 14. Server: The emotion engine recognizes emotions from the other party's voice and generates "calm" emotion data.

[1980] 15. Server: The translation engine translates the text into Japanese as "My name is John," maintaining a calm tone.

[1981] 16. Server: The speech synthesis engine converts Japanese text into speech data.

[1982] 17. Server: Sends audio data to the user's terminal.

[1983] 18. Device: Play the audio data "My name is John." in a calm tone.

[1984] In this way, smooth and considerate communication between the user and the person speaking the foreign language is achieved.

[1985] The following describes the processing flow.

[1986] Step 1:

[1987] User (device): The user speaks to their smartphone or tablet and says, "Hello, what's your name?"

[1988] Step 2:

[1989] Terminal: The voice input application records the user's speech and converts it into a digital format.

[1990] Step 3:

[1991] Terminal: The audio data, converted to digital format, is stored locally, and that data is sent to a cloud-based speech recognition API.

[1992] Step 4:

[1993] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "Hello, what is your name?".

[1994] Step 5:

[1995] Server: Send text data to the emotion engine and analyze the user's emotion.

[1996] Step 6:

[1997] Server: The emotion engine analyzes the tone and pitch of the voice, etc., and recognizes that the user has a "happy" emotion. Also, generate emotion data.

[1998] Step 7:

[1999] Server: Send the voice-recognized text data and emotion data to the translation engine.

[2000] Step 8:

[2001] Server: The translation engine translates the text data "Konnichiwa, onamae wa nan desu ka?" into English as "Hello, what is your name?", and adjusts the expression to match the user's "happy" emotion.

[2002] Step 9:

[2003] Server: Send the translated English text data "Hello, what is your name?" to the speech synthesis engine.

[2004] Step 10:

[2005] Server: The speech synthesis engine converts the English text data into voice data.

[2006] Step 11:

[2007] Server: Send the generated voice data to the terminal.

[2008] Step 12:

[2009] Terminal: Receives audio data and plays it back to the user as audio. For example, it might play a cheerful voice saying, "Hello, what is your name?"

[2010] Step 13:

[2011] The other party (device): The other party responds with "My name is John."

[2012] Step 14:

[2013] Recipient's device: Records the recipient's speech and converts it to a digital format.

[2014] Step 15:

[2015] The recipient's device: It saves the audio data, converted to digital format, locally and sends that data to a cloud-based speech recognition API.

[2016] Step 16:

[2017] Server: A cloud-based speech recognition API receives the audio data and converts it into text data. This text data is "My name is John."

[2018] Step 17:

[2019] Server: Sends text data to the emotion engine to analyze the other party's emotions.

[2020] Step 18:

[2021] Server: The emotion engine analyzes the tone and pitch of the voice and recognizes that the other party has a "calm" emotion. It also generates emotion data.

[2022] Step 19:

[2023] Server: Transmits the voice - recognized text data and emotion data to the translation engine.

[2024] Step 20:

[2025] Server: The translation engine translates the text data "My name is John." into Japanese as "私の名前はジョンです。" and adjusts the expression to fit the "calm" emotion of the other party.

[2026] Step 21:

[2027] Server: Transmits the translated Japanese text data "私の名前はジョンです。" to the speech synthesis engine.

[2028] Step 22:

[2029] Server: The speech synthesis engine converts the Japanese text data into speech data.

[2030] Step 23:

[2031] Server: Transmits the generated speech data to the user's terminal.

[2032] Step 24:

[2033] Terminal: Receives the speech data and plays it as speech to the user. For example, the speech "私の名前はジョンです。" is played in a calm tone.

[2034] In this way, smooth communication considering emotions is realized between the user and the foreign - language speaker.

[2035] (Example 2)

[2036] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the robot 414 is referred to as the "terminal".

[2037] In recent years, international communication has increased, and smooth multilingual conversation is required. However, conventional translation systems simply translate languages ​​without adjusting the translation to take into account the speaker's emotions, resulting in a decline in the quality of communication. Furthermore, in real-time speech translation, the processing speed from speech recognition to translation and playback has been a problem. Therefore, there is a need for a system that accurately reflects the speaker's emotions while enabling smooth, real-time communication between multiple languages.

[2038] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[2039] In this invention, the server includes means for recognizing emotions from the user's voice in real time and generating emotion data, means for adjusting translated expressions based on translated text and emotion data, and means for converting text translated into a foreign language into speech data. This enables real-time multilingual voice communication using translated expressions that reflect the speaker's emotions.

[2040] A "user" refers to a person who uses the system to perform voice input and voice translation.

[2041] "Digital format" refers to a format in which analog audio is converted into digital data that can be processed by a computer.

[2042] "Speech recognition means" refers to means that have the function of converting digital speech data into text data.

[2043] "Translation means" refers to a means that has the function of converting text from one language to another language.

[2044] "Emotion recognition means" refers to means that have the function of analyzing the speaker's emotions in real time from audio data and generating emotion data.

[2045] "Means for adjusting translated expressions" refers to means that have the function of adjusting translated expressions to reflect appropriate sentiment based on the translated text and sentiment data.

[2046] "Speech synthesis means" refers to means that have the function of converting text data into speech data.

[2047] "Playback means" refers to means that have the function of outputting the generated audio data as sound.

[2048] A "cloud-based speech recognition engine" refers to an engine that uses servers and software accessible via a network to convert speech data into text data.

[2049] "Communication equipment" refers to electronic devices with communication capabilities, such as smartphones and tablets.

[2050] "Audio equipment" refers to devices such as speakers and earphones used to play audio data.

[2051] This invention is a speech translation system that enables users to communicate smoothly with people who speak a foreign language. The system consists of the following processes: speech input, speech recognition, emotion recognition, text translation, speech synthesis, and speech playback. Each process utilizes specific hardware and software.

[2052] System Configuration

[2053] 1. Voice input

[2054] User: Speaks into a smartphone or tablet. For example, says, "Hello, what's your name?"

[2055] Terminal: Uses a microphone as an interface to convert the user's voice into a digital format.

[2056] 2. Speech Recognition

[2057] Terminal: Sends digital voice data to a cloud-based speech recognition engine. The API used is the speech recognition engine.

[2058] Server: Converts the digital audio data received by the speech recognition engine into text data.

[2059] 3. Emotion recognition

[2060] Server: The emotion recognition engine analyzes the user's voice in real time and generates emotion data. The emotion recognition engine used is an emotion analysis engine.

[2061] 4. Text Translation

[2062] Server: Sends the speech-recognized text data and generated sentiment data to the translation engine. The translation engine used is a text translation engine.

[2063] Server: The translation engine converts the text into the target language and adjusts the translation based on sentiment data.

[2064] 5. Speech synthesis

[2065] Server: Sends the translated text data to the speech synthesis engine. The speech synthesis engine used is speech synthesis software.

[2066] Server: The speech synthesis engine converts text data into speech data.

[2067] 6. Audio Playback

[2068] Terminal: Receives audio data and plays it back using the terminal's sound device. For example, it might play back "Hello, what is your name?" in a cheerful tone.

[2069] Specific example

[2070] As a concrete example, we will simulate a conversation between a Japanese-speaking user and an English-speaking partner. The following is the processing flow.

[2071] 1. User: "Hello, what's your name?" they say to their smartphone.

[2072] 2. Device: Records audio and converts it into digital data.

[2073] 3. Server: The speech recognition engine converts "Hello, what is your name?" into text data.

[2074] 4. Server: The emotion recognition engine generates text data and emotion data, and detects the emotion of "happiness".

[2075] 5. Server: The translation engine translates the text data into English as "Hello, what is your name?", maintaining a cheerful tone.

[2076] 6. Server: The speech synthesis engine converts the translated text into speech data.

[2077] 7. Terminal: Receives audio data and plays "Hello, what is your name?" in a cheerful tone.

[2078] Example of a prompt

[2079] Use the following prompts to input data into the generative AI model.

[2080] "Please translate the Japanese phrase 'Hello, what is your name?' into English and convert it into speech, while maintaining the emotion in the speech."

[2081] "Translate the English sentence 'My name is John.' into Japanese and then speak it in a calm tone."

[2082] In this way, real-time voice translation that takes emotions into consideration is achieved between the user and the person speaking the foreign language.

[2083] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2084] Program processing flow

[2085] Step 1: Voice Input

[2086] User: Speak into the microphone on your smartphone or tablet and say, "Hello, what's your name?"

[2087] Input: Analog audio

[2088] Output: Digital audio data

[2089] Terminal: Captures the user's voice with a microphone, converts the analog audio to a digital format using an audio digitization module, and saves it to a buffer.

[2090] Step 2: Speech Recognition

[2091] Terminal: Sends digital voice data to a cloud-based speech recognition engine.

[2092] Input: Digital audio data

[2093] Output: Text data ("Hello, what is your name?")

[2094] Server: The speech recognition engine analyzes the digital speech data it receives and converts it into text data, taking into account phonemes and context.

[2095] Step 3: Emotion Recognition

[2096] Server: Sends the speech-recognized text data and digital audio data to the emotion recognition engine.

[2097] Input: Text data, digital audio data

[2098] Output: Emotional data (e.g., "Happy")

[2099] Server: The emotion recognition engine analyzes the tone, pitch, and speed of the voice, evaluates the user's emotional state in real time, and generates emotion data.

[2100] Step 4: Text Translation

[2101] Server: Sends the speech-recognized text data and sentiment data to the translation engine.

[2102] Input: Text data, sentiment data

[2103] Output: Translated text data (e.g., "Hello, what is your name?")

[2104] Server: The translation engine converts text data into the target foreign language and adjusts the translation based on sentiment data.

[2105] Step 5: Speech Synthesis

[2106] Server: Sends the translated text data to the speech synthesis engine.

[2107] Input: Translated text data

[2108] Output: Audio data

[2109] Server: The speech synthesis engine analyzes the translated text data and generates speech data that reflects emotional tone.

[2110] Step 6: Audio Playback

[2111] Device: Receives audio data and plays it through the device's speaker.

[2112] Input: Audio data

[2113] Output: Played audio (e.g., "Hello, what is your name?")

[2114] Terminal: Uses an audio device to play the generated audio data to the user.

[2115] (Application Example 2)

[2116] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2117] In physical stores, communication between customers who speak different languages ​​and employees often breaks down. This can lead to decreased customer satisfaction and a decline in the quality of service. Furthermore, it is difficult for employees to immediately understand and respond to multiple languages, especially when it comes to conveying nuances, including emotions. In this situation, it is necessary to improve customer service in physical stores by streamlining foreign language support and providing accurate translations that take emotions into consideration.

[2118] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[2119] This invention includes a server that includes means for inputting a user's voice and converting the voice into a digital format, speech recognition means for converting the digitally converted voice into text, translation means for translating the text into a predetermined foreign language, an emotion engine that recognizes the emotion of the translated text and adjusts the translated expression appropriately according to the emotion, speech synthesis means for converting the translated text into audio data, playback means for playing the audio data back to the user, and a device that implements an application that supports real-time multilingual communication between the user and customers who speak a foreign language. This makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the emotions of customers.

[2120] "Means of inputting audio and converting it to a digital format" refers to devices or processes that receive audio signals and record them as digital data.

[2121] "Speech recognition means" refers to a technology or device that analyzes speech converted into digital data and converts it into corresponding text data.

[2122] "Translation methods" refer to software or algorithms used to convert text data from one language into another language.

[2123] An "emotion engine" is a technology or algorithm that recognizes a speaker's emotions from audio or text data and analyzes that emotional information.

[2124] "Speech synthesis means" refers to a technology or device that analyzes text data and generates speech data corresponding to that text.

[2125] "Reproduction means" refers to a device or technology used to actually pronounce the generated audio data as sound.

[2126] "Device" is a general term for any equipment on which an application is installed, such as communication terminals, smartphones, and smart glasses.

[2127] A "cloud-based speech recognition engine" refers to a service or technology for performing speech recognition in a cloud computing environment.

[2128] A "communication terminal" refers to an electronic device such as a smartphone or tablet that has functions such as voice input and playback, and data communication.

[2129] An "application that supports multilingual communication" is software designed to enable users who speak different languages ​​to communicate with each other in real time.

[2130] This invention is a system that facilitates multilingual support in physical stores and provides appropriate translations that take emotions into consideration. This system uses a device to support real-time communication between users and foreign language-speaking customers. The main elements of the system are means for inputting voice and converting it to a digital format, voice recognition means, translation means, emotion engine, voice synthesis means, playback means, and a device for implementing the application.

[2131] The following hardware and software will be used to implement this system:

[2132] Hardware: Communication devices (e.g., smartphones, tablets, smart glasses)

[2133] software:

[2134] Speech recognition: Google Cloud Speech-to-Text API

[2135] Translation: Google Cloud Translation API

[2136] Emotion recognition: IBM Watson Tone Analyzer API

[2137] Text-to-speech: Google Cloud Text-to-Speech API

[2138] Real-time communication: Firebase Realtime Database

[2139] Description of the system's program processing

[2140] First, the user speaks into a communication device. The communication device converts this audio into a digital format and uploads it to Firebase. The server sends the audio data stored in Firebase to the Google Cloud Speech-to-Text API, where it converts the audio into text data.

[2141] Next, the server sends this text data to the IBM Watson Tone Analyzer API for sentiment analysis. The text, including sentiment data, is then sent to the Google Cloud Translation API for translation into the specified foreign language. Based on the sentiment data, the translated sentence is adjusted to an appropriate tone.

[2142] The server then sends the translated text data to the Google Cloud Text-to-Speech API to generate audio data. This audio data is then stored again in Firebase and sent to the user's communication device. The communication device plays this audio data, allowing the user to provide the translated message to their customers.

[2143] Specific example

[2144] Let's consider a situation where an employee working in a souvenir shop in a tourist area needs to communicate smoothly with foreign customers. For example,

[2145] User (employee): "Welcome, how can I help you?" they say to their smartphone.

[2146] System: The server recognizes the speech, analyzes the emotions, translates it into English, and plays back "Welcome! How can I help you?" in a friendly tone.

[2147] The other party (customer) asks, "Can you tell me where I can find local souvenirs?"

[2148] User: Provide a properly translated response through the system again to facilitate smooth communication.

[2149] Example of a prompt

[2150] "Based on the following information, please generate code that performs translations tailored to the customer's emotions."

[2151] Customer comment: "Hello, can I purchase this?"

[2152] Emotion: "Friendly"

[2153] Translated statement: "Hello, can I purchase this?"

[2154] Code for speech synthesis in a friendly tone

[2155] This invention makes it possible to provide smooth multilingual support in physical stores and to achieve smooth communication while taking into consideration the customer's feelings.

[2156] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2157] Step 1:

[2158] Input: The user speaks into the communication terminal.

[2159] Operation: The user speaks to the communication terminal saying, "Welcome, how can I help you?"

[2160] Data processing: The communication terminal converts this audio into a digital format.

[2161] Output: Digital audio data is generated.

[2162] Step 2:

[2163] Input: Digital audio data.

[2164] Operation: The communication device uploads this digital audio data to Firebase.

[2165] Data processing: Digital audio data is transferred to the cloud.

[2166] Output: Digital audio data is stored on Firebase.

[2167] Step 3:

[2168] Input: Digital audio data stored in Firebase.

[2169] Operation: The server retrieves audio data from Firebase and sends it to the Google Cloud Speech-to-Text API.

[2170] Data processing: The speech recognition engine analyzes the digital audio data and converts it into text data.

[2171] Output: The text data "Welcome, how can I help you?" is generated.

[2172] Step 4:

[2173] Input: Text data.

[2174] Operation: The server sends text data to the IBM Watson Tone Analyzer API.

[2175] Data processing: The emotion engine analyzes the text data and generates emotion data (e.g., "friendly tone").

[2176] Output: Sentiment data is generated and attached to the text data.

[2177] Step 5:

[2178] Input: Text data and sentiment data.

[2179] Operation: The server sends this data to the Google Cloud Translation API.

[2180] Data processing: The translation engine translates text data into the specified foreign language and adjusts it based on sentiment data.

[2181] Output: The translated text data "Welcome! How can I help you?" is generated.

[2182] Step 6:

[2183] Input: Translated text data.

[2184] Operation: The server sends this text data to the Google Cloud Text-to-Speech API.

[2185] Data processing: The speech synthesis engine converts the translated text data into speech data.

[2186] Output: The audio data "Welcome! How can I help you?" is generated.

[2187] Step 7:

[2188] Input: Translated audio data.

[2189] Operation: The server uploads the generated audio data to Firebase and sends it to the user's communication device.

[2190] Data processing: Audio data is transferred from the server to the communication terminal.

[2191] Output: The audio data is saved to the communication terminal.

[2192] Step 8:

[2193] Input: Translated audio data.

[2194] Operation: The communication terminal plays the translated audio data through its speaker.

[2195] Data processing: Audio data is output as audio.

[2196] Output: The translated audio "Welcome! How can I help you?" is played to the customer.

[2197] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[2198] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2199] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[2200] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2201] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[2202] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[2203] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[2204] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[2205] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[2206] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[2207] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[2208] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[2209] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[2210] 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.

[2211] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[2212] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[2213] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[2214] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[2215] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[2216] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[2217] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[2218] The following is further disclosed regarding the embodiments described above.

[2219] (Claim 1)

[2220] A means for inputting user voice and converting said voice into a digital format,

[2221] A speech recognition means that converts the audio converted to the aforementioned digital format into text,

[2222] A translation means for translating the aforementioned text into a predetermined foreign language,

[2223] A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data,

[2224] A system including a playback means for playing the aforementioned audio data to a user.

[2225] (Claim 2)

[2226] The system according to claim 1, wherein the speech recognition means transmits the speech converted to digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

[2227] (Claim 3)

[2228] The system according to claim 1, wherein the playback means is a speaker implemented in the user's smartphone or tablet.

[2229] "Example 1"

[2230] (Claim 1)

[2231] A means for inputting user voice and converting said voice into a digital format,

[2232] A speech recognition means that converts the audio converted to the aforementioned digital format into text,

[2233] A translation means for translating the aforementioned text into a predetermined foreign language,

[2234] A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data,

[2235] A playback means for playing the aforementioned audio data to the user,

[2236] Means for transmitting text data obtained from the speech recognition means to a translation engine,

[2237] Means for transmitting the translated text obtained from the translation engine to a speech synthesis engine,

[2238] Means for transmitting the voice data obtained from the speech synthesis engine to the user's terminal,

[2239] A system that includes this.

[2240] (Claim 2)

[2241] The system according to claim 1, wherein the speech recognition means transmits the speech converted to digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

[2242] (Claim 3)

[2243] The system according to claim 1, wherein the playback means is an audio output device implemented in the user's mobile information terminal.

[2244] "Application Example 1"

[2245] (Claim 1)

[2246] A means for inputting user voice and converting said voice into a digital format,

[2247] A speech recognition means that converts the audio converted to the aforementioned digital format into text,

[2248] A translation means for translating the aforementioned text into a predetermined foreign language,

[2249] A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data,

[2250] A playback means for playing the aforementioned audio data to the user,

[2251] A means to support communication between customers and employees within the store,

[2252] A system that includes means for customers or employees to speak into a device.

[2253] (Claim 2)

[2254] The system according to claim 1, wherein the speech recognition means transmits the speech converted to digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

[2255] (Claim 3)

[2256] The system according to claim 1, wherein the playback means is a speaker installed in the user's mobile information terminal.

[2257] "Example 2 of combining an emotion engine"

[2258] (Claim 1)

[2259] A means for inputting user voice and converting said voice into a digital format,

[2260] A speech recognition means that converts the audio converted to the aforementioned digital format into text,

[2261] A translation means for translating the aforementioned text into a predetermined foreign language,

[2262] An emotion recognition means that recognizes the user's emotions in real time from the aforementioned audio and generates emotion data,

[2263] Means for adjusting the translated expression based on the translated text and sentiment data,

[2264] A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data,

[2265] A system including a playback means for playing the aforementioned audio data to a user.

[2266] (Claim 2)

[2267] The system according to claim 1, wherein the speech recognition means transmits the speech converted to digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

[2268] (Claim 3)

[2269] The system according to claim 1, wherein the playback means is an acoustic device implemented in the user's communication device.

[2270] "Application example 2 when combining with an emotional engine"

[2271] (Claim 1)

[2272] A means for inputting user voice and converting said voice into a digital format,

[2273] A speech recognition means that converts the audio converted to the aforementioned digital format into text,

[2274] A translation means for translating the aforementioned text into a predetermined foreign language,

[2275] An emotion engine that recognizes the emotion of the translated text and adjusts the translated expression appropriately according to that emotion,

[2276] A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data,

[2277] A playback means for playing the aforementioned audio data to the user,

[2278] A device that implements an application to support real-time multilingual communication between the aforementioned user and customers who speak foreign languages.

[2279] A system that includes this.

[2280] (Claim 2)

[2281] The system according to claim 1, wherein the speech recognition means transmits the speech converted to digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

[2282] (Claim 3)

[2283] The system according to claim 1, wherein the playback means is a speaker installed in the user's communication terminal. [Explanation of Symbols]

[2284] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for inputting user voice and converting said voice into a digital format, A speech recognition means that converts the audio converted to the aforementioned digital format into text, A translation means for translating the aforementioned text into a predetermined foreign language, A speech synthesis means that converts the text translated into the aforementioned foreign language into audio data, A system including a playback means for playing the aforementioned audio data to a user.

2. The system according to claim 1, wherein the speech recognition means transmits the speech converted to the digital format to a cloud-based speech recognition engine, and the cloud-based speech recognition engine converts the speech to text.

3. The system according to claim 1, wherein the playback means is a speaker installed in the user's smartphone or tablet.

Images