System

Abstract

To provide a system for achieving real communication experience between a user and a character.SOLUTION: A communication system in which a user and a character can interact with each other in real time, comprising: means for acquiring a voice input from the user; means for converting the voice input into text data; means for analyzing the text data and generating an appropriate response; means for converting the generated response from the text data into voice data; and means for outputting the voice data to the user.SELECTED DRAWING: Figure 11

Description

[Technical Field]

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

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

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

[0004] The problem this invention aims to solve is to provide an environment in which users can have natural conversations with anime and game characters in real time in the real world. Conventional dialogue systems only allow limited responses and preset dialogues, resulting in a poor user experience. Furthermore, these systems make it difficult to provide new business opportunities for companies that hold character rights. The objective of this invention is to improve this issue and realize a realistic communication experience between users and characters. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means.

[0006] The present invention provides a communication system including a means for receiving voice input from a user, a means for converting the voice input into text data, a means for analyzing the text data and generating an appropriate response, a means for converting the generated response from the text data into voice data, and a means for outputting the voice data to the user. The system also includes a means for learning character-specific lines and voice information, and can provide dialogue with different characters according to user selection. This allows users to have an experience that feels as if they are actually conversing with the characters, and also provides new business opportunities for character rights holders.

[0007] "User" refers to an individual who uses the system and interacts with the characters.

[0008] "Voice input" refers to the process by which the system obtains voice information spoken by the user.

[0009] "Text data" refers to data that has been analyzed and converted into text information from voice input.

[0010] "Analysis" refers to the process of understanding the meaning and intent of text data.

[0011] "Response" refers to a response from a character generated based on the analysis results.

[0012] "Audio data" refers to information that has been converted from text data into audio format.

[0013] "Output" refers to the process of sending and playing audio data to the user.

[0014] "Communication System" refers to a system that includes a set of processing devices and software that allows a user and a character to interact in real time.

[0015] "Character" refers to personified beings such as people and animals that appear in fiction such as anime and games.

[0016] "Lines" refers to the words or phrases spoken by a character.

[0017] "Audio information" refers to audio data uttered by a specific character.

[0018] "Learning" refers to the process by which a system improves the accuracy of its model based on specific data.

[0019] "Selection" refers to the act of a user specifying the character they wish to use.

[0020] "Dialogue" refers to two-way communication between the user and the character. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0022] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0023] First, the terms used in the following description will be explained.

[0024] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0025] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0026] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0027] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0028] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0032] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0033] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0034] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0035] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0036] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0039] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0040] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0041] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0042] This invention relates to a communication system that allows users to have natural conversations with characters in anime and games in real time. The program processing of this system will be specifically described below.

[0043] server

[0044] 1. Collecting and Learning Character Information:

[0045] The server collects dialogue and voice data provided by character rights holders, and the collected data is pre-processed to remove noise and check grammar.

[0046] Based on the pre-processed data, the generative AI model learns the characters' unique lines and personalities, and the learning results are stored in a database.

[0047] 2. Generating and sending conversation data:

[0048] The server receives voice input from the user's device, which is converted into text data by a speech recognition system.

[0049] The converted text data is analyzed using natural language processing (NLP) and processed to understand the user's intent.

[0050] Based on the analysis results, an AI model is used to generate an appropriate response, which is generated as text data.

[0051] Finally, the generated text data is sent to the terminal.

[0052] Terminal

[0053] 1. Capture voice input and convert it to text:

[0054] The device receives the user's voice input and captures the voice data through the microphone, which is then sent to a voice recognition system and converted into text data.

[0055] 2. Sending a conversation request and receiving a response:

[0056] The converted text data is sent to the server and a response from the server is awaited.

[0057] The terminal receives the response data returned from the server.

[0058] 3. Transcribing and playing back response data:

[0059] The received text data is sent to a speech synthesis system and converted into voice data.

[0060] The generated audio data is played back through the figure's speaker so that the user can hear it.

[0061] user

[0062] 1. Character selection and conversation start:

[0063] Users use a terminal application to select their favorite character, and this information is stored on the terminal.

[0064] When you speak to the selected character, the figure will begin a conversation.

[0065] 2. Receiving the response:

[0066] The user receives a voice response from the figure, and repeats the process if they wish to continue the conversation.

[0067] Specific examples

[0068] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[0069] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[0070] The processing flow will be explained below.

[0071] server

[0072] Step 1:

[0073] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[0074] Step 2:

[0075] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[0076] Step 3:

[0077] The preprocessed data is analyzed and the speech is converted into text data using a speech recognition system.

[0078] Step 4:

[0079] The text data is used to further train the generative AI model, which learns dialogue based on information about the character's personality and behavior.

[0080] Step 5:

[0081] The learning results are stored in a database and can be accessed immediately upon user request.

[0082] Step 6:

[0083] The user sends voice input via the terminal to the server, which receives it and converts it into text data using a voice recognition system.

[0084] Step 7:

[0085] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intentions and emotions.

[0086] Step 8:

[0087] Based on the analysis results, a generative AI model is used to generate an appropriate response, which is then generated as text data.

[0088] Step 9:

[0089] The generated text data is sent to the terminal and a response is returned to the user.

[0090] Terminal

[0091] Step 1:

[0092] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[0093] Step 2:

[0094] The captured voice data is sent to a voice recognition system and converted into text data.

[0095] Step 3:

[0096] Sends the converted text data to the server and waits for a response from the server.

[0097] Step 4:

[0098] Receives response data from the server in text format.

[0099] Step 5:

[0100] The received text data is sent to a speech synthesis system and converted into voice data.

[0101] Step 6:

[0102] The generated audio data is played through the figure's speaker, and the user hears an audible response from the figure.

[0103] user

[0104] Step 1:

[0105] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[0106] Step 2:

[0107] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[0108] Step 3:

[0109] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[0110] Step 4:

[0111] The user receives a voice response from the figure and continues the conversation. By repeating the same steps, the user can continue to have natural conversations with the character.

[0112] Through these processing steps, the system enables realistic and natural interactions between users and characters. By having each entity work together, it is possible to improve the quality of the user experience and provide new business opportunities for character rights holders.

[0113] Example 1

[0114] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0115] In systems that allow users to have natural conversations with anime and game characters in real time, existing technologies have had difficulty adapting character-specific lines and personalities in real time. Furthermore, when changing characters according to user selection, the technology lacked the flexibility to provide different conversation patterns for each character.

[0116] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0117] In this invention, the server includes: [means for acquiring voice input from a user;] [means for converting the voice input into text data;] [means for analyzing the text data and generating an appropriate response;] [means for learning the character's unique lines and personality using a generative AI model;] [means for converting the generated response from text data into voice data; and [means for outputting the voice data to the user.] This allows the user to enjoy natural conversations in real time with the selected character.

[0118] "Means for obtaining voice input from a user" means devices and software for capturing the user's speech as digital voice data.

[0119] "Means for converting voice input into text data" refers to software and algorithms for converting captured voice data into text information using voice recognition technology.

[0120] "Means for analyzing text data and generating appropriate responses" refers to software and algorithms that analyze the converted text data using natural language processing technology, understand the user's intent, and generate appropriate text responses.

[0121] "Means for using a generative AI model to learn character-specific lines and personality traits" refers to a technology that uses provided character data to train a machine learning model to learn a character's specific speech patterns and personality traits.

[0122] The "means for converting the generated response from text data to speech data" is software and algorithms for converting the generated text data into natural-sounding speech data using text-to-speech (TTS) technology.

[0123] "Means for outputting audio data to a user" refers to equipment and software for playing the generated audio data through a speaker or other audio output device so that the user can hear it.

[0124] This invention relates to a communication system that allows users to have natural conversations with anime and game characters in real time. The system utilizes a generative AI model to convert the user's voice input into text data, analyze the text data, and generate appropriate responses.

[0125] server

[0126] 1. Collecting and Learning Character Information:

[0127] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed, which includes noise removal and grammar checks. Based on the preprocessed data, a generative AI model (e.g., a general-purpose machine learning model) learns the character's unique dialogue and personality. The learning results are stored in a database.

[0128] 2. Receiving voice input and converting it to text:

[0129] The server receives voice input from the user's device and converts the received voice data into text data using a voice recognition system (e.g., a general voice recognition API).

[0130] 3. Analyzing user intent:

[0131] The server then analyzes the converted text data using natural language processing (NLP) techniques, which provides the basis for understanding the user's intent and generating an appropriate response.

[0132] 4. Generate response data:

[0133] The server generates an appropriate response based on the parsed text data, using a generative AI model, and sends the generated text data to the device as a response.

[0134] Terminal

[0135] 1. Acquiring voice input:

[0136] The device captures the user's voice input through the microphone, and the captured voice data is temporarily stored in a buffer on the device.

[0137] 2. Send a request for text conversion:

[0138] The device sends the stored voice data to the server and makes a request to convert the voice to text.

[0139] 3. Receiving a response from the server:

[0140] The terminal receives the response data sent from the server, and this received data is temporarily stored in the terminal's memory.

[0141] 4. Speech synthesis and playback:

[0142] The device sends the received text data to a speech synthesis system (e.g., general speech synthesis software) and converts it into voice data, which is then played back through the figure's speaker.

[0143] user

[0144] 1. Character Selection:

[0145] Users select their favorite character using the device application, and this information is stored on the device and also sent to the server.

[0146] 2. Voice input:

[0147] Users can speak to the character they choose, for example, by saying, "How was your day?"

[0148] 3. Receiving the response:

[0149] The user hears a response from the figure, such as "Today was a fun day."

[0150] Specific examples

[0151] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[0152] Prompt Sentence Examples

[0153] "Design a system that allows a character figure to interact with the user in a natural way. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[0154] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[0155] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0156] Step 1:

[0157] The user selects a character. The user operates the application on the device and selects the character they like. This information is stored on the device and also sent to the server. The input is the user's selection, and the output is the selected character information.

[0158] Step 2:

[0159] The user provides voice input. The user speaks what they want to say to the character into the microphone. The input is the user's voice, and the output is saved as audio data on the device.

[0160] Step 3:

[0161] The device captures audio input. The device uses a microphone to capture the user's voice as digital audio data and stores it in a buffer. The input is the user's voice, and the output is digital audio data.

[0162] Step 4:

[0163] The device sends voice data to the server. The device sends saved voice data to the server and makes a voice recognition request. The input is voice data, and the output is voice data sent to the server.

[0164] Step 5:

[0165] The server converts the voice data into text data. The server converts the voice data into text information using a voice recognition system (for example, a general voice recognition API). The input is voice data, and the output is text data.

[0166] Step 6:

[0167] The server analyzes the text data. The server uses natural language processing (NLP) technology to analyze the text data and understand the user's intent. The input is the text data, and the output is the analysis result.

[0168] Step 7:

[0169] The server generates an appropriate response. The server uses a generative AI model to generate a response based on the analysis results. This response is generated as text data. The input is the analysis results, and the output is the generated text data.

[0170] Step 8:

[0171] The server sends the generated response to the terminal. The server sends the generated text data to the terminal. The input is the generated text data, and the output is the text data sent to the terminal.

[0172] Step 9:

[0173] The device converts text data into speech data. The device uses a speech synthesis system (e.g., common speech synthesis software) to convert text data into speech data. The input is text data and the output is speech data.

[0174] Step 10:

[0175] The terminal plays the generated audio data through the figure's speaker. The input is the audio data and the output is the played audio.

[0176] Example prompt sentence:

[0177] "Design a system that allows natural dialogue between a character figure and the user. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[0178] (Application example 1)

[0179] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0180] In virtual stores, the experience for users to obtain product information has not been intuitive enough, and effective natural dialogue has not been realized. Conventional systems lack the technology to enable natural dialogue between users and characters, resulting in a poor user experience. Therefore, there is a need for a method to enable users to efficiently obtain product information and improve satisfaction through more natural, real-time dialogue.

[0181] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0182] In this invention, the server includes: [means for acquiring voice input from a user]; [means for converting the voice input into text data]; [means for analyzing the text data and generating an appropriate response]; [means for converting the generated response from text data into voice data]; [means for outputting the voice data to the user]; and [means for a character in a virtual store to provide product information and improve the user's interaction experience.] This enables the user to obtain product information through natural interaction with the character in the virtual store.

[0183] "Capturing user voice input" is the process of capturing voice input data from a user and incorporating it into the system.

[0184] "Speech-to-text" is the process of converting captured voice data into text data.

[0185] "Text data analysis and response generation" is the process of analyzing the converted text data, understanding the user's intent, and generating an appropriate response.

[0186] "Conversion to voice data" refers to the process of converting the text data of the generated response into voice data.

[0187] "Outputting audio data" is the process of making the converted audio data audible to the user through a device such as a speaker.

[0188] "Learning character-specific lines and voice information" is the process of learning lines and voice characteristics specific to a particular character and using them to conduct natural conversations.

[0189] "Providing dialogue with different characters" is a process in which different characters are used depending on the user's selection, and the characters provide dialogue with the user.

[0190] "Providing product information in a virtual store" is the process by which a character in a virtual environment provides information about products to users.

[0191] "Improving the user's interaction experience" is a process aimed at qualitatively improving the experience that users have through natural interactions with characters.

[0192] This program is a system that allows users to obtain product information through natural dialogue with characters in a virtual store and assists the purchasing process. The system is composed of a server, a terminal, and user operations.

[0193] server

[0194] The server has the following responsibilities:

[0195] 1. Collecting and Learning Character Information:

[0196] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed to remove noise and check grammar. Based on this preprocessed data, the generative AI model learns the character's unique dialogue and personality. The learning results are stored in a database.

[0197] 2. Generating and sending conversation data:

[0198] The server receives voice input from the device and converts it into text using a speech recognition system. The converted text is then analyzed using natural language processing (NLP) to understand the user's intent and generate an appropriate response, which is then sent back to the device.

[0199] Terminal

[0200] The terminal has the following responsibilities:

[0201] 1. Capture voice input and convert it to text:

[0202] It takes the user's voice input and captures it through a microphone, which sends the voice data to a speech recognition system and converts it into text data.

[0203] 2. Sending a conversation request and receiving a response:

[0204] The converted text data is sent to the server, and a response is waited for from the server. The terminal again receives the response data sent back from the server.

[0205] 3. Transcribing and playing back response data:

[0206] The received text data is converted into voice data using a speech synthesis system, and the generated voice data is played back through the device's speaker, allowing the user to listen to it.

[0207] user

[0208] The user operates the following functions:

[0209] 1. Character selection and conversation start:

[0210] Users use a terminal application to select their favorite character, which is stored on the terminal, and can then start a conversation by speaking to the selected character.

[0211] 2. Receiving the response:

[0212] The user receives a voice response from the character, and if they wish to continue the conversation, they repeat the same process.

[0213] Hardware and software used

[0214] Hardware: smartphone, microphone, speaker

[0215] Software: Python, SpeechRecognition library, Transformers library, gTTS (Google Text-to-Speech)

[0216] Specific examples

[0217] When a user asks the figure, "What products do you recommend?", the following process takes place:

[0218] The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "The recommended product is the latest smartphone," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figure replies, "The recommended product is the latest smartphone." In this way, the user enjoys a realistic dialogue with the character.

[0219] Prompt Sentence Examples

[0220] 1. "What products do you recommend?"

[0221] 2. "Please tell me the features of this product."

[0222] 3. "Do you have it in stock?"

[0223] 4. "When is the next sale?"

[0224] This allows the system to enable users to obtain product information through natural interactions with characters in the virtual store, improving the user experience.

[0225] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0226] Step 1:

[0227] The user speaks into the device's microphone, and the user's voice input is captured through the device's microphone.

[0228] Input: User's voice

[0229] Output: Audio data

[0230] Step 2:

[0231] The device sends the captured voice data to a speech recognition system that converts it into text using the SpeechRecognition library.

[0232] Input: Audio data

[0233] Output: Text data

[0234] Step 3:

[0235] The text data is transmitted from the terminal to the server.

[0236] Input: Text data

[0237] Output: Text data sent to the server

[0238] Step 4:

[0239] The server analyzes the received text data using natural language processing (NLP). Specifically, a generative AI model using the Transformers library analyzes the text data and understands the user's intent.

[0240] Input: Text data sent to the server

[0241] Output: Analysis results

[0242] Step 5:

[0243] The server generates an appropriate response based on the analysis results, using a generative AI model to generate the best response to the user's question or request.

[0244] Input: Analysis results

[0245] Output: Response text data

[0246] Step 6:

[0247] The generated response text data is transmitted from the server to the terminal.

[0248] Input: Response text data

[0249] Output: Response text data sent to the terminal

[0250] Step 7:

[0251] The device sends the received response text data to a speech synthesis system, which converts it into voice data. The speech synthesis system uses gTTS (Google Text-to-Speech).

[0252] Input: Response text data sent to the terminal

[0253] Output: Response audio data

[0254] Step 8:

[0255] The generated audio data is played back through the device's speaker, allowing the user to hear it.

[0256] Input: Response audio data

[0257] Output: The response the user hears

[0258] To summarize the specific operations of each step, the user speaks a question into the device, the speech is converted into text and sent to the server, the server analyzes the text data, generates an appropriate response and sends it back to the device, and the device converts the received text response into speech and plays it back to the user, achieving natural dialogue in real time.

[0259] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0260] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with characters in anime or games in real time. The program processing of this system will be specifically described below.

[0261] server

[0262] 1. Collecting and Learning Character Information:

[0263] The server collects dialogue and voice data provided by character rights holders. This data is pre-processed to remove unwanted noise. Based on the pre-processed data, the generative AI model learns the character's unique dialogue and personality.

[0264] 2. Receiving and parsing voice input from the user:

[0265] When a user makes a voice input, the data is sent via the terminal to the server, which receives the voice data and converts it into text data using a voice recognition system.

[0266] 3. Sentiment Analysis of Text Data:

[0267] The server sends the converted text data to the emotion engine, which analyzes the user's emotions and determines their emotional state from the tone and content of their voice.

[0268] 4. Generating an appropriate response:

[0269] Based on the results of the emotion analysis, the server uses a generative AI model to generate an appropriate response. For example, if the user looks sad, the character will respond with comforting words.

[0270] 5. Sending response data:

[0271] The generated response data is sent to the terminal and a reply is sent to the user.

[0272] Terminal

[0273] 1. Capture voice input and convert it to text:

[0274] The device uses a microphone to capture the user's voice, which is then sent to a voice recognition system and converted into text.

[0275] 2. Sending the converted data and receiving the response:

[0276] The terminal transmits the converted text data to the server, waits for response data from the server, and receives the text data returned from the server.

[0277] 3. Transcribing and playing back response data:

[0278] The device sends the received text data to a speech synthesis system, converts it into voice data, and plays the generated voice data through the figure's speaker to respond to the user.

[0279] user

[0280] 1. Character selection and conversation start:

[0281] Users use the terminal application to select their favorite character, and when they speak to the selected character, the figure captures their voice.

[0282] 2. Receiving responses and reflecting emotions:

[0283] The figure responds with a voice response that is based on the user's emotional state, and the emotion engine adjusts the response to take the user's state into account, allowing for more natural and meaningful interactions.

[0284] Specific examples

[0285] For example, if a user says, "Today was not a very fun day," the following process occurs: The device captures the voice data, converts it into text data, and sends it to the server. The server analyzes the text data with an emotion engine and determines that the user is sad. The generative AI model generates comforting words such as "What happened? Thanks for telling me," and sends the text data back to the device. The device converts this data into speech, and the figure replies, "What happened? Thanks for telling me." In this way, the user can continue the dialogue with the character in a natural way.

[0286] Through these processing steps, the system achieves realistic and natural dialogue between the user and the character. The incorporation of an emotion engine generates appropriate responses based on the user's emotions, further improving the quality of the user experience.

[0287] The processing flow will be explained below.

[0288] server

[0289] Step 1:

[0290] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[0291] Step 2:

[0292] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[0293] Step 3:

[0294] Analyze the pre-processed data and convert the speech into text data. Use a speech recognition system to efficiently convert the collected speech information into text.

[0295] Step 4:

[0296] The text data is used to further train the generative AI model, which then incorporates the character's unique lines and personality traits and updates its parameters.

[0297] Step 5:

[0298] The learning results are stored in a database to enable quick responses to user requests.

[0299] Step 6:

[0300] When a user sends a voice input to the server via the terminal, the server receives it and converts the voice data into text data using a voice recognition system.

[0301] Step 7:

[0302] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intent.

[0303] Step 8:

[0304] Analyze user sentiment from text data using an emotion engine, which identifies emotions from the user's phrasing and tone of voice.

[0305] Step 9:

[0306] Based on the analysis results, an appropriate response is generated using a generative AI model, which is then structured as text data.

[0307] Step 10:

[0308] The generated text data is sent to the terminal and a reply is given to the user, which is tailored to the user's emotions.

[0309] Terminal

[0310] Step 1:

[0311] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[0312] Step 2:

[0313] The captured voice data is sent to a voice recognition system and converted into text data using voice recognition software.

[0314] Step 3:

[0315] The converted text data is sent to the server and a response is awaited from the server, a necessary step to prepare the server for processing the data.

[0316] Step 4:

[0317] Receives the response data received from the server in text format. This data was generated by the server based on the user's request.

[0318] Step 5:

[0319] The received text data is sent to a speech synthesis system and converted into voice data. Through the speech synthesis process, the text data is turned into voice.

[0320] Step 6:

[0321] The generated voice data is played through the figure's speaker, and the figure uses this voice data to respond to the user.

[0322] user

[0323] Step 1:

[0324] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[0325] Step 2:

[0326] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[0327] Step 3:

[0328] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[0329] Step 4:

[0330] The user can continue the conversation by receiving voice responses from the figure. The responses correspond to the user's emotions, allowing for natural dialogue.

[0331] Examples:

[0332] For example, if a user says to a figurine, "You seem tired and down lately," the following process takes place: The device captures the voice data and converts it into text data using a voice recognition system. This text data is sent to the server, where an emotion engine analyzes the emotion "tired." Based on the analysis results, the generative AI model generates a response such as "Are you OK? You should get some rest," and sends it back to the device as text data. The device converts this data into voice, and the figurine responds, "Are you OK? You should get some rest." In this way, the user can continue to have a natural conversation with the character.

[0333] Through these specific processing steps, the system achieves realistic and natural dialogue between users and characters. The built-in emotion engine generates appropriate responses based on the user's emotions, further improving the user experience and providing new business opportunities for character rights holders.

[0334] Example 2

[0335] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0336] Conventional communication systems are unable to generate appropriate responses based on the user's emotional state, resulting in conversations that tend to be one-sided and mechanical. Furthermore, they lack the means to learn character-specific lines and personalities, resulting in a poor user experience. There is a need for a communication system that can solve these issues and realize more natural and emotionally sensitive conversations.

[0337] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0338] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data with an emotion analysis engine and identifying the user's emotional state, means for generating an appropriate response using a generative AI model based on the results of the emotion analysis, means for converting the generated response from text data into voice data, and means for outputting the voice data to the user, thereby enabling natural and meaningful dialogue that is in tune with the user's emotional state.

[0339] "Means for obtaining voice input from a user" refers to a device or method for collecting voice data uttered by a user using a microphone or voice input device.

[0340] The "means for converting speech input into text data" refers to a device or method that uses a speech recognition system to convert acquired speech data into text information.

[0341] "Means for analyzing text data using an emotion analysis engine and identifying a user's emotional state" refers to a device or method that uses text data as input and analyzes and identifies a user's emotional state (e.g., joy, sadness, anger, etc.).

[0342] "Means for generating an appropriate response using a generative AI model based on the results of emotion analysis" refers to a device or method that uses a generative AI model to create a natural and appropriate response, taking into account the user's emotional state obtained through emotion analysis.

[0343] The "means for converting the generated response from text data to voice data" refers to a device or method for converting a response expressed in text data into voice data using a voice synthesis system.

[0344] "Means for outputting audio data to a user" means a device or method for communicating the generated audio data to a user through a speaker or other audio output device.

[0345] "Means for learning character-specific lines and personality information" refers to the process of using a machine learning model to learn given lines and personality information to reflect the character's personality and speech.

[0346] A "means for providing dialogue with different characters in response to a user's selection" is a device or method that generates responses and dialogue specific to the character selected by the user.

[0347] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with anime and game characters in real time. This system is realized by the roles of a server, terminals, and users, and their cooperation.

[0348] Server Features

[0349] The server collects character information, analyzes voice input, analyzes emotions, generates responses, and transmits response data.

[0350] 1. Collecting and Learning Character Information:

[0351] The server collects dialogue and voice data provided by character rights holders. This data is preprocessed using Python and LibROSA to remove noise. The preprocessed data is then trained by a generative AI model (e.g., GPT-3). This training process allows the system to analyze the character's unique dialogue and personality, resulting in natural-sounding dialogue.

[0352] 2. Receive and analyze user voice input:

[0353] The user's speech is captured by the device and sent to a server, which converts the speech data into text using a speech recognition system, such as Google Speech-to-Text.

[0354] 3. Emotion analysis:

[0355] The server then sends the converted text data to IBM Watson's emotion analysis engine to analyze the user's emotional state. Through this process, the user's emotions (such as joy, sadness, anger, etc.) are identified.

[0356] 4. Generate response:

[0357] Based on the results of the emotion analysis, the generative AI model generates an appropriate response. For example, if the analysis finds that the user is sad, the generative AI model generates comforting words such as "What happened? Thank you for telling me." An example of a prompt sentence could be, "If the user is sad, please generate comforting words."

[0358] 5. Sending response data:

[0359] The server sends the generated response text data to the terminal, which converts this data into voice data and outputs it to the user.

[0360] Device Features

[0361] The terminal is responsible for acquiring user voice, converting it into text data, communicating with the server, and outputting response data as voice.

[0362] 1. Capture voice input and convert it to text:

[0363] The device uses a microphone to capture the user's voice, which is then converted into text data using, for example, Google Speech-to-Text.

[0364] 2. Communication with the server:

[0365] The converted text data is sent to the server, and at the same time, a process is performed to wait for and receive response data from the server.

[0366] 3. Transcribing and playing back response data:

[0367] The device receives the text data returned from the server and converts it into voice data using a speech synthesis system such as Amazon Polly, which then responds to the user through the speaker.

[0368] User Actions

[0369] Users can select a character using a terminal application and begin interacting with that character. The figure's responses are adjusted according to the user's emotional state, allowing for more natural and rich interactions.

[0370] 1. Character selection and dialogue initiation:

[0371] Users select their favorite character through the terminal application, and then speak to the character to activate the dialogue system.

[0372] 2. Receiving responses and reflecting emotions:

[0373] You can receive voice responses from the figure and enjoy natural dialogue that reflects your own emotional state.

[0374] Specific examples

[0375] For example, if a user says, "Today was not a very fun day," the device captures the voice data and converts it into text using Google Speech-to-Text. The server analyzes the text data using IBM Watson's emotion analysis engine and determines that the user is sad. The generative AI model generates a response, "What happened? Thank you for talking," based on the prompt, "If the user is sad, please generate words of comfort." The device converts this response into speech using Amazon Polly, and the figure replies, "What happened? Thank you for talking."

[0376] Through this process, the system achieves realistic and natural dialogue between the user and the character. Appropriate responses generated based on the analysis results of the emotion engine further improve the quality of the dialogue.

[0377] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0378] Step 1: Gather and learn about the character

[0379] The server collects dialogue and voice data provided by character rights holders. The voice data is preprocessed and noise-removed using Python and LibROSA. The input is the provided character's dialogue and voice data, and the output is clean voice data with noise removed.

[0380] Specific operation: The server loads the collected audio files with LibROSA, performs noise filtering using FFT (Fast Fourier Transform), and generates a preprocessed dataset.

[0381] Step 2: Training the generative AI model

[0382] The server trains a generative AI model (e.g., GPT-3) based on the preprocessed data. The input is the preprocessed dataset, and the output is a model that has learned the character's unique lines and personality.

[0383] How it works: The server uses the dataset to train the AI ​​model, building a model that can generate responses that reflect the character's lines and personality traits.

[0384] Step 3: Getting voice input

[0385] The device captures the user's voice as they speak into the microphone. The input is the user's voice data, and the output is the captured raw audio file.

[0386] What it does: The device's microphone captures the user's speech and records it in real time.

[0387] Step 4: Speech to text

[0388] The device sends the captured voice data to a speech recognition system such as Google Speech-to-Text and converts it into text data. The input is the user's voice data, and the output is the corresponding text data.

[0389] Specific operation: The device sends the audio file to the cloud API and receives the returned text data.

[0390] Step 5: Sentiment analysis of text data

[0391] The server receives the text data sent from the device and performs emotion analysis using IBM Watson's emotion analysis engine. The input is the text data, and the output is the analyzed emotional state.

[0392] Specific operation: The server runs the text data through a sentiment analysis engine, resulting in the user's emotion (e.g., joy, sadness, anger).

[0393] Step 6: Generate a response

[0394] The server generates an appropriate response using a generative AI model based on the results of the sentiment analysis. The input is the sentiment analysis result and a prompt, and the output is a generated text response. An example of a prompt is "If the user is sad, please generate comforting words."

[0395] Specific operation: The server inputs a prompt to the generative AI model: "If the user is sad, please generate words of comfort." The model generates a response text such as "What happened? Thank you for telling me."

[0396] Step 7: Transcribing response data

[0397] The device sends the text data sent from the server to a speech synthesis system such as Amazon Polly and converts it into voice data. The input is the response text data, and the output is the synthesized voice data.

[0398] Specific operation: The device sends text data to the speech synthesis API and receives the generated audio file.

[0399] Step 8: Outputting audio data

[0400] The device outputs voice data to the user through a speaker. The input is synthesized voice data, and the output is the voice played from the speaker.

[0401] What happens: The device speaker plays the audio file and the user listens to it.

[0402] Through these processing steps, the system enables natural and emotionally responsive interactions between the user and the character.

[0403] (Application example 2)

[0404] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0405] In conventional communication systems, the dialogue between the user and the character is mechanical and one-way, rather than emotionally responsive. Therefore, there is a need to realize more natural and meaningful dialogue. Furthermore, if appropriate product suggestions can be made based on the user's emotional state, the customer experience can be improved.

[0406] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice input from the user, means for converting the voice input into text data, means for recognizing emotions based on the analyzed text data, and means for generating a response and making product suggestions based on the user's emotional state. This makes the dialogue between the user and the character natural and meaningful according to emotions, and further enables appropriate product suggestions based on the user's emotional state.

[0407] A "user" is a person who uses this system to interact with characters.

[0408] "Voice input" refers to the voice data that a user speaks to the system through a microphone.

[0409] "Text data" refers to character data converted by a voice recognition system based on voice input.

[0410] "Analysis" refers to the process of analyzing text data using an emotion recognition engine or similar to understand the user's emotions and intentions.

[0411] "Emotion recognition" is the process of identifying a user's emotional state through analysis.

[0412] "Generating an appropriate response" is the process of generating a reply to establish a dialogue based on the user's emotional state.

[0413] "Product suggestion" is the act of recommending appropriate products based on the user's emotional state and the content of their conversation.

[0414] "Voice data" refers to data that has been converted back into voice format from text data using a voice synthesis system.

[0415] "Output" refers to the act of transmitting the generated audio data to the user through a speaker.

[0416] A "system" is a collection of hardware and software necessary to realize interaction between a user and a character.

[0417] The system for implementing this invention utilizes emotion analysis data to provide appropriate responses and product recommendations in a virtual store where users can interact with anime and game characters. The detailed configuration of this system and the method for realizing it are described below.

[0418] System Configuration

[0419] 1. Server

[0420] Character information gathering and learning:

[0421] The server collects and preprocesses the dialogue and voice data provided by the character rights holders. After preprocessing such as noise removal, the generative AI model learns the character's unique dialogue and personality.

[0422] Receiving and parsing voice input:

[0423] Users input their voice into the virtual store, which is then received by the server, where the voice data is converted into text data using a voice recognition system.

[0424] Sentiment analysis of text data:

[0425] The server sends the analyzed text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) to identify the user's emotion.

[0426] Generate appropriate responses and product suggestions:

[0427] Based on the results of the emotion analysis, a generative AI model (e.g., OpenAI GPT-3) is used to generate appropriate responses, and in addition, product suggestions are made based on the user's emotional state.

[0428] Sending response data:

[0429] The generated response and product proposal are transmitted to the terminal as text data.

[0430] 2. Terminal

[0431] Capture voice input and convert it to text:

[0432] Voice data is captured by the user's microphone and converted into text data using a voice recognition system (e.g., Google Speech-to-Text).

[0433] Sending the transformed data and receiving the response:

[0434] The converted text data is sent to the server, and response data from the server is waited for and received.

[0435] Transcribing and playing response data:

[0436] The received text data is sent to a speech synthesis system (e.g., Amazon Polly) and converted into voice data. The generated voice data is played back through the speaker to respond to the user.

[0437] 3. Users

[0438] Character selection and conversation initiation:

[0439] Users can use the virtual store application, select their favorite character, and start interacting with them. When they speak, the microphone captures the sound and sends it to the device.

[0440] Receiving responses and reflecting emotions:

[0441] The device receives voice feedback responses, which are tailored to the user's emotional state based on an emotion recognition engine.

[0442] Examples and prompts

[0443] For example, if a user says "I'm not feeling well today" in a virtual store, the following happens:

[0444] 1. The device captures the voice data, converts it into text data, and sends it to the server.

[0445] 2. The server performs speech recognition and emotion analysis and determines that the text data "I'm not feeling well" represents a sad emotion.

[0446] 3. The server's generated AI model generates data suggesting products to lift the mood (e.g., pleasant-smelling aromas) along with a response such as, "How's your day? Tell us if you'd like."

[0447] 4. The response data and product suggestions are sent to the terminal and converted into voice by a voice synthesis system.

[0448] 5. The user is prompted with the message, "How's your day going? Let me know if you'd like." along with suggestions for aroma products.

[0449] Prompt Sentence Examples

[0450] "You are a character in a virtual store. A user says, 'I'm feeling down today.' You encourage the user and suggest products that will lift their spirits."

[0451] In this way, this system provides natural and meaningful responses and product suggestions that correspond to the user's emotions through dialogue between the user and the character.

[0452] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0453] Step 1:

[0454] Users simply launch the virtual store application and select their favorite character, which then appears on the screen and can begin interacting with it.

[0455] Step 2:

[0456] The user speaks to the character, saying, "I'm not feeling well today." The device captures this voice data through the microphone and receives it as input.

[0457] Step 3:

[0458] The device uses a voice recognition system (e.g., Google Speech-to-Text) to convert the acquired voice data into text data. At this time, the device processes the voice recognition API and obtains the text data as output.

[0459] Step 4:

[0460] The terminal sends the obtained text data to the server. The text data is transferred as input to the server via the network, and the server receives it.

[0461] Step 5:

[0462] The server sends the received text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) for emotion analysis. The text data is passed as input to the analysis engine, and the analyzed user's emotional state is obtained as output.

[0463] Step 6:

[0464] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-3) based on the results of the emotion analysis. At this time, the prompt sentence is given as input to the generative AI model, and response text data that matches the user's emotion is obtained as output.

[0465] Step 7:

[0466] In addition to the generated responses, the server also generates product suggestions based on the user's emotional state. Using the emotion analysis results as input, the server uses a generative AI model to generate text data containing product suggestions as output.

[0467] Step 8:

[0468] The server transmits text data including the response and the product proposal to the terminal, and transmits the text data generated as input to the terminal as output via the network.

[0469] Step 9:

[0470] The device sends the received text data to a speech synthesis system (e.g., Amazon Polly) and converts it into voice data. Text data is passed as input to the speech synthesis API, and the generated voice data is obtained as output.

[0471] Step 10:

[0472] The generated voice data is played from the device's speaker to deliver a response to the user. Specifically, the device outputs the generated voice data to the speaker, and the user listens to it.

[0473] Step 11:

[0474] The user receives responses and suggestions from the character, and decides whether to continue the conversation or consider a product, thus creating a natural and emotionally relevant continuous interaction between the user and the character.

[0475] Through the above processing steps, the user can have a rich interaction experience with the character and can also receive product suggestions according to their emotional state.

[0476] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0477] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0478] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0479] [Second embodiment]

[0480] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0481] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0482] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0483] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0484] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0485] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0486] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0487] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0488] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0489] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0490] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0491] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0492] This invention relates to a communication system that allows users to have natural conversations with characters in anime and games in real time. The program processing of this system will be specifically described below.

[0493] server

[0494] 1. Collecting and Learning Character Information:

[0495] The server collects dialogue and voice data provided by character rights holders, and the collected data is pre-processed to remove noise and check grammar.

[0496] Based on the pre-processed data, the generative AI model learns the characters' unique lines and personalities, and the learning results are stored in a database.

[0497] 2. Generating and sending conversation data:

[0498] The server receives voice input from the user's device, which is converted into text data by a speech recognition system.

[0499] The converted text data is analyzed using natural language processing (NLP) and processed to understand the user's intent.

[0500] Based on the analysis results, an AI model is used to generate an appropriate response, which is generated as text data.

[0501] Finally, the generated text data is sent to the terminal.

[0502] Terminal

[0503] 1. Capture voice input and convert it to text:

[0504] The device receives the user's voice input and captures the voice data through the microphone, which is then sent to a voice recognition system and converted into text data.

[0505] 2. Sending a conversation request and receiving a response:

[0506] The converted text data is sent to the server and a response from the server is awaited.

[0507] The terminal receives the response data returned from the server.

[0508] 3. Transcribing and playing back response data:

[0509] The received text data is sent to a speech synthesis system and converted into voice data.

[0510] The generated audio data is played back through the figure's speaker so that the user can hear it.

[0511] user

[0512] 1. Character selection and conversation start:

[0513] Users use a terminal application to select their favorite character, and this information is stored on the terminal.

[0514] When you speak to the selected character, the figure will begin a conversation.

[0515] 2. Receiving the response:

[0516] The user receives a voice response from the figure, and repeats the process if they wish to continue the conversation.

[0517] Specific examples

[0518] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[0519] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[0520] The processing flow will be explained below.

[0521] server

[0522] Step 1:

[0523] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[0524] Step 2:

[0525] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[0526] Step 3:

[0527] The preprocessed data is analyzed and the speech is converted into text data using a speech recognition system.

[0528] Step 4:

[0529] The text data is used to further train the generative AI model, which learns dialogue based on information about the character's personality and behavior.

[0530] Step 5:

[0531] The learning results are stored in a database and can be accessed immediately upon user request.

[0532] Step 6:

[0533] The user sends voice input via the terminal to the server, which receives it and converts it into text data using a voice recognition system.

[0534] Step 7:

[0535] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intentions and emotions.

[0536] Step 8:

[0537] Based on the analysis results, a generative AI model is used to generate an appropriate response, which is then generated as text data.

[0538] Step 9:

[0539] The generated text data is sent to the terminal and a response is returned to the user.

[0540] Terminal

[0541] Step 1:

[0542] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[0543] Step 2:

[0544] The captured voice data is sent to a voice recognition system and converted into text data.

[0545] Step 3:

[0546] Sends the converted text data to the server and waits for a response from the server.

[0547] Step 4:

[0548] Receives response data from the server in text format.

[0549] Step 5:

[0550] The received text data is sent to a speech synthesis system and converted into voice data.

[0551] Step 6:

[0552] The generated audio data is played through the figure's speaker, and the user hears an audible response from the figure.

[0553] user

[0554] Step 1:

[0555] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[0556] Step 2:

[0557] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[0558] Step 3:

[0559] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[0560] Step 4:

[0561] The user receives a voice response from the figure and continues the conversation. By repeating the same steps, the user can continue to have natural conversations with the character.

[0562] Through these processing steps, the system enables realistic and natural interactions between users and characters. By having each entity work together, it is possible to improve the quality of the user experience and provide new business opportunities for character rights holders.

[0563] Example 1

[0564] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0565] In systems that allow users to have natural conversations with anime and game characters in real time, existing technologies have had difficulty adapting character-specific lines and personalities in real time. Furthermore, when changing characters according to user selection, the technology lacked the flexibility to provide different conversation patterns for each character.

[0566] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0567] In this invention, the server includes: [means for acquiring voice input from a user;] [means for converting the voice input into text data;] [means for analyzing the text data and generating an appropriate response;] [means for learning the character's unique lines and personality using a generative AI model;] [means for converting the generated response from text data into voice data; and [means for outputting the voice data to the user.] This allows the user to enjoy natural conversations in real time with the selected character.

[0568] "Means for obtaining voice input from a user" means devices and software for capturing the user's speech as digital voice data.

[0569] "Means for converting voice input into text data" refers to software and algorithms for converting captured voice data into text information using voice recognition technology.

[0570] "Means for analyzing text data and generating appropriate responses" refers to software and algorithms that analyze the converted text data using natural language processing technology, understand the user's intent, and generate appropriate text responses.

[0571] "Means for using a generative AI model to learn character-specific lines and personality traits" refers to a technology that uses provided character data to train a machine learning model to learn a character's specific speech patterns and personality traits.

[0572] The "means for converting the generated response from text data to speech data" is software and algorithms for converting the generated text data into natural-sounding speech data using text-to-speech (TTS) technology.

[0573] "Means for outputting audio data to a user" refers to equipment and software for playing the generated audio data through a speaker or other audio output device so that the user can hear it.

[0574] This invention relates to a communication system that allows users to have natural conversations with anime and game characters in real time. The system utilizes a generative AI model to convert the user's voice input into text data, analyze the text data, and generate appropriate responses.

[0575] server

[0576] 1. Collecting and Learning Character Information:

[0577] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed, which includes noise removal and grammar checks. Based on the preprocessed data, a generative AI model (e.g., a general-purpose machine learning model) learns the character's unique dialogue and personality. The learning results are stored in a database.

[0578] 2. Receiving voice input and converting it to text:

[0579] The server receives voice input from the user's device and converts the received voice data into text data using a voice recognition system (e.g., a general voice recognition API).

[0580] 3. Analyzing user intent:

[0581] The server then analyzes the converted text data using natural language processing (NLP) techniques, which provides the basis for understanding the user's intent and generating an appropriate response.

[0582] 4. Generate response data:

[0583] The server generates an appropriate response based on the parsed text data, using a generative AI model, and sends the generated text data to the device as a response.

[0584] Terminal

[0585] 1. Acquiring voice input:

[0586] The device captures the user's voice input through the microphone, and the captured voice data is temporarily stored in a buffer on the device.

[0587] 2. Send a request for text conversion:

[0588] The device sends the stored voice data to the server and makes a request to convert the voice to text.

[0589] 3. Receiving a response from the server:

[0590] The terminal receives the response data sent from the server, and this received data is temporarily stored in the terminal's memory.

[0591] 4. Speech synthesis and playback:

[0592] The device sends the received text data to a speech synthesis system (e.g., general speech synthesis software) and converts it into voice data, which is then played back through the figure's speaker.

[0593] user

[0594] 1. Character Selection:

[0595] Users select their favorite character using the device application, and this information is stored on the device and also sent to the server.

[0596] 2. Voice input:

[0597] Users can speak to the character they choose, for example, by saying, "How was your day?"

[0598] 3. Receiving the response:

[0599] The user hears a response from the figure, such as "Today was a fun day."

[0600] Specific examples

[0601] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[0602] Prompt Sentence Examples

[0603] "Design a system that allows a character figure to interact with the user in a natural way. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[0604] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[0605] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0606] Step 1:

[0607] The user selects a character. The user operates the application on the device and selects the character they like. This information is stored on the device and also sent to the server. The input is the user's selection, and the output is the selected character information.

[0608] Step 2:

[0609] The user provides voice input. The user speaks what they want to say to the character into the microphone. The input is the user's voice, and the output is saved as audio data on the device.

[0610] Step 3:

[0611] The device captures audio input. The device uses a microphone to capture the user's voice as digital audio data and stores it in a buffer. The input is the user's voice, and the output is digital audio data.

[0612] Step 4:

[0613] The device sends voice data to the server. The device sends saved voice data to the server and makes a voice recognition request. The input is voice data, and the output is voice data sent to the server.

[0614] Step 5:

[0615] The server converts the voice data into text data. The server converts the voice data into text information using a voice recognition system (for example, a general voice recognition API). The input is voice data, and the output is text data.

[0616] Step 6:

[0617] The server analyzes the text data. The server uses natural language processing (NLP) technology to analyze the text data and understand the user's intent. The input is the text data, and the output is the analysis result.

[0618] Step 7:

[0619] The server generates an appropriate response. The server uses a generative AI model to generate a response based on the analysis results. This response is generated as text data. The input is the analysis results, and the output is the generated text data.

[0620] Step 8:

[0621] The server sends the generated response to the terminal. The server sends the generated text data to the terminal. The input is the generated text data, and the output is the text data sent to the terminal.

[0622] Step 9:

[0623] The device converts text data into speech data. The device uses a speech synthesis system (e.g., common speech synthesis software) to convert text data into speech data. The input is text data and the output is speech data.

[0624] Step 10:

[0625] The terminal plays the generated audio data through the figure's speaker. The input is the audio data and the output is the played audio.

[0626] Example prompt sentence:

[0627] "Design a system that allows natural dialogue between a character figure and the user. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[0628] (Application example 1)

[0629] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0630] In virtual stores, the experience for users to obtain product information has not been intuitive enough, and effective natural dialogue has not been realized. Conventional systems lack the technology to enable natural dialogue between users and characters, resulting in a poor user experience. Therefore, there is a need for a method to enable users to efficiently obtain product information and improve satisfaction through more natural, real-time dialogue.

[0631] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0632] In this invention, the server includes: [means for acquiring voice input from a user]; [means for converting the voice input into text data]; [means for analyzing the text data and generating an appropriate response]; [means for converting the generated response from text data into voice data]; [means for outputting the voice data to the user]; and [means for a character in a virtual store to provide product information and improve the user's interaction experience.] This enables the user to obtain product information through natural interaction with the character in the virtual store.

[0633] "Capturing user voice input" is the process of capturing voice input data from a user and incorporating it into the system.

[0634] "Speech-to-text" is the process of converting captured voice data into text data.

[0635] "Text data analysis and response generation" is the process of analyzing the converted text data, understanding the user's intent, and generating an appropriate response.

[0636] "Conversion to voice data" refers to the process of converting the text data of the generated response into voice data.

[0637] "Outputting audio data" is the process of making the converted audio data audible to the user through a device such as a speaker.

[0638] "Learning character-specific lines and voice information" is the process of learning lines and voice characteristics specific to a particular character and using them to conduct natural conversations.

[0639] "Providing dialogue with different characters" is a process in which different characters are used depending on the user's selection, and the characters provide dialogue with the user.

[0640] "Providing product information in a virtual store" is the process by which a character in a virtual environment provides information about products to users.

[0641] "Improving the user's interaction experience" is a process aimed at qualitatively improving the experience that users have through natural interactions with characters.

[0642] This program is a system that allows users to obtain product information through natural dialogue with characters in a virtual store and assists the purchasing process. The system is composed of a server, a terminal, and user operations.

[0643] server

[0644] The server has the following responsibilities:

[0645] 1. Collecting and Learning Character Information:

[0646] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed to remove noise and check grammar. Based on this preprocessed data, the generative AI model learns the character's unique dialogue and personality. The learning results are stored in a database.

[0647] 2. Generating and sending conversation data:

[0648] The server receives voice input from the device and converts it into text using a speech recognition system. The converted text is then analyzed using natural language processing (NLP) to understand the user's intent and generate an appropriate response, which is then sent back to the device.

[0649] Terminal

[0650] The terminal has the following responsibilities:

[0651] 1. Capture voice input and convert it to text:

[0652] It takes the user's voice input and captures it through a microphone, which sends the voice data to a speech recognition system and converts it into text data.

[0653] 2. Sending a conversation request and receiving a response:

[0654] The converted text data is sent to the server, and a response is waited for from the server. The terminal again receives the response data sent back from the server.

[0655] 3. Transcribing and playing back response data:

[0656] The received text data is converted into voice data using a speech synthesis system, and the generated voice data is played back through the device's speaker, allowing the user to listen to it.

[0657] user

[0658] The user operates the following functions:

[0659] 1. Character selection and conversation start:

[0660] Users use a terminal application to select their favorite character, which is stored on the terminal, and can then start a conversation by speaking to the selected character.

[0661] 2. Receiving the response:

[0662] The user receives a voice response from the character, and if they wish to continue the conversation, they repeat the same process.

[0663] Hardware and software used

[0664] Hardware: smartphone, microphone, speaker

[0665] Software: Python, SpeechRecognition library, Transformers library, gTTS (Google Text-to-Speech)

[0666] Specific examples

[0667] When a user asks the figure, "What products do you recommend?", the following process takes place:

[0668] The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "The recommended product is the latest smartphone," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figure replies, "The recommended product is the latest smartphone." In this way, the user enjoys a realistic dialogue with the character.

[0669] Prompt Sentence Examples

[0670] 1. "What products do you recommend?"

[0671] 2. "Please tell me the features of this product."

[0672] 3. "Do you have it in stock?"

[0673] 4. "When is the next sale?"

[0674] This allows the system to enable users to obtain product information through natural interactions with characters in the virtual store, improving the user experience.

[0675] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0676] Step 1:

[0677] The user speaks into the device's microphone, and the user's voice input is captured through the device's microphone.

[0678] Input: User's voice

[0679] Output: Audio data

[0680] Step 2:

[0681] The device sends the captured voice data to a speech recognition system that converts it into text using the SpeechRecognition library.

[0682] Input: Audio data

[0683] Output: Text data

[0684] Step 3:

[0685] The text data is transmitted from the terminal to the server.

[0686] Input: Text data

[0687] Output: Text data sent to the server

[0688] Step 4:

[0689] The server analyzes the received text data using natural language processing (NLP). Specifically, a generative AI model using the Transformers library analyzes the text data and understands the user's intent.

[0690] Input: Text data sent to the server

[0691] Output: Analysis results

[0692] Step 5:

[0693] The server generates an appropriate response based on the analysis results, using a generative AI model to generate the best response to the user's question or request.

[0694] Input: Analysis results

[0695] Output: Response text data

[0696] Step 6:

[0697] The generated response text data is transmitted from the server to the terminal.

[0698] Input: Response text data

[0699] Output: Response text data sent to the terminal

[0700] Step 7:

[0701] The device sends the received response text data to a speech synthesis system, which converts it into voice data. The speech synthesis system uses gTTS (Google Text-to-Speech).

[0702] Input: Response text data sent to the terminal

[0703] Output: Response audio data

[0704] Step 8:

[0705] The generated audio data is played back through the device's speaker, allowing the user to hear it.

[0706] Input: Response audio data

[0707] Output: The response the user hears

[0708] To summarize the specific operations of each step, the user speaks a question into the device, the speech is converted into text and sent to the server, the server analyzes the text data, generates an appropriate response and sends it back to the device, and the device converts the received text response into speech and plays it back to the user, achieving natural dialogue in real time.

[0709] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0710] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with characters in anime or games in real time. The program processing of this system will be specifically described below.

[0711] server

[0712] 1. Collecting and Learning Character Information:

[0713] The server collects dialogue and voice data provided by character rights holders. This data is pre-processed to remove unwanted noise. Based on the pre-processed data, the generative AI model learns the character's unique dialogue and personality.

[0714] 2. Receiving and parsing voice input from the user:

[0715] When a user makes a voice input, the data is sent via the terminal to the server, which receives the voice data and converts it into text data using a voice recognition system.

[0716] 3. Sentiment Analysis of Text Data:

[0717] The server sends the converted text data to the emotion engine, which analyzes the user's emotions and determines their emotional state from the tone and content of their voice.

[0718] 4. Generating an appropriate response:

[0719] Based on the results of the emotion analysis, the server uses a generative AI model to generate an appropriate response. For example, if the user looks sad, the character will respond with comforting words.

[0720] 5. Sending response data:

[0721] The generated response data is sent to the terminal and a reply is sent to the user.

[0722] Terminal

[0723] 1. Capture voice input and convert it to text:

[0724] The device uses a microphone to capture the user's voice, which is then sent to a voice recognition system and converted into text.

[0725] 2. Sending the converted data and receiving the response:

[0726] The terminal transmits the converted text data to the server, waits for response data from the server, and receives the text data returned from the server.

[0727] 3. Transcribing and playing back response data:

[0728] The device sends the received text data to a speech synthesis system, converts it into voice data, and plays the generated voice data through the figure's speaker to respond to the user.

[0729] user

[0730] 1. Character selection and conversation start:

[0731] Users use the terminal application to select their favorite character, and when they speak to the selected character, the figure captures their voice.

[0732] 2. Receiving responses and reflecting emotions:

[0733] The figure responds with a voice response that is based on the user's emotional state, and the emotion engine adjusts the response to take the user's state into account, allowing for more natural and meaningful interactions.

[0734] Specific examples

[0735] For example, if a user says, "Today was not a very fun day," the following process occurs: The device captures the voice data, converts it into text data, and sends it to the server. The server analyzes the text data with an emotion engine and determines that the user is sad. The generative AI model generates comforting words such as "What happened? Thanks for telling me," and sends the text data back to the device. The device converts this data into speech, and the figure replies, "What happened? Thanks for telling me." In this way, the user can continue the dialogue with the character in a natural way.

[0736] Through these processing steps, the system achieves realistic and natural dialogue between the user and the character. The incorporation of an emotion engine generates appropriate responses based on the user's emotions, further improving the quality of the user experience.

[0737] The processing flow will be explained below.

[0738] server

[0739] Step 1:

[0740] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[0741] Step 2:

[0742] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[0743] Step 3:

[0744] Analyze the pre-processed data and convert the speech into text data. Use a speech recognition system to efficiently convert the collected speech information into text.

[0745] Step 4:

[0746] The text data is used to further train the generative AI model, which then incorporates the character's unique lines and personality traits and updates its parameters.

[0747] Step 5:

[0748] The learning results are stored in a database to enable quick responses to user requests.

[0749] Step 6:

[0750] When a user sends a voice input to the server via the terminal, the server receives it and converts the voice data into text data using a voice recognition system.

[0751] Step 7:

[0752] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intent.

[0753] Step 8:

[0754] Analyze user sentiment from text data using an emotion engine, which identifies emotions from the user's phrasing and tone of voice.

[0755] Step 9:

[0756] Based on the analysis results, an appropriate response is generated using a generative AI model, which is then structured as text data.

[0757] Step 10:

[0758] The generated text data is sent to the terminal and a reply is given to the user, which is tailored to the user's emotions.

[0759] Terminal

[0760] Step 1:

[0761] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[0762] Step 2:

[0763] The captured voice data is sent to a voice recognition system and converted into text data using voice recognition software.

[0764] Step 3:

[0765] The converted text data is sent to the server and a response is awaited from the server, a necessary step to prepare the server for processing the data.

[0766] Step 4:

[0767] Receives the response data received from the server in text format. This data was generated by the server based on the user's request.

[0768] Step 5:

[0769] The received text data is sent to a speech synthesis system and converted into voice data. Through the speech synthesis process, the text data is turned into voice.

[0770] Step 6:

[0771] The generated voice data is played through the figure's speaker, and the figure uses this voice data to respond to the user.

[0772] user

[0773] Step 1:

[0774] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[0775] Step 2:

[0776] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[0777] Step 3:

[0778] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[0779] Step 4:

[0780] The user can continue the conversation by receiving voice responses from the figure. The responses correspond to the user's emotions, allowing for natural dialogue.

[0781] Examples:

[0782] For example, if a user says to a figurine, "You seem tired and down lately," the following process takes place: The device captures the voice data and converts it into text data using a voice recognition system. This text data is sent to the server, where an emotion engine analyzes the emotion "tired." Based on the analysis results, the generative AI model generates a response such as "Are you OK? You should get some rest," and sends it back to the device as text data. The device converts this data into voice, and the figurine responds, "Are you OK? You should get some rest." In this way, the user can continue to have a natural conversation with the character.

[0783] Through these specific processing steps, the system achieves realistic and natural dialogue between users and characters. The built-in emotion engine generates appropriate responses based on the user's emotions, further improving the user experience and providing new business opportunities for character rights holders.

[0784] Example 2

[0785] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0786] Conventional communication systems are unable to generate appropriate responses based on the user's emotional state, resulting in conversations that tend to be one-sided and mechanical. Furthermore, they lack the means to learn character-specific lines and personalities, resulting in a poor user experience. There is a need for a communication system that can solve these issues and realize more natural and emotionally sensitive conversations.

[0787] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0788] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data with an emotion analysis engine and identifying the user's emotional state, means for generating an appropriate response using a generative AI model based on the results of the emotion analysis, means for converting the generated response from text data into voice data, and means for outputting the voice data to the user, thereby enabling natural and meaningful dialogue that is in tune with the user's emotional state.

[0789] "Means for obtaining voice input from a user" refers to a device or method for collecting voice data uttered by a user using a microphone or voice input device.

[0790] The "means for converting speech input into text data" refers to a device or method that uses a speech recognition system to convert acquired speech data into text information.

[0791] "Means for analyzing text data using an emotion analysis engine and identifying a user's emotional state" refers to a device or method that uses text data as input and analyzes and identifies a user's emotional state (e.g., joy, sadness, anger, etc.).

[0792] "Means for generating an appropriate response using a generative AI model based on the results of emotion analysis" refers to a device or method that uses a generative AI model to create a natural and appropriate response, taking into account the user's emotional state obtained through emotion analysis.

[0793] The "means for converting the generated response from text data to voice data" refers to a device or method for converting a response expressed in text data into voice data using a voice synthesis system.

[0794] "Means for outputting audio data to a user" means a device or method for communicating the generated audio data to a user through a speaker or other audio output device.

[0795] "Means for learning character-specific lines and personality information" refers to the process of using a machine learning model to learn given lines and personality information to reflect the character's personality and speech.

[0796] A "means for providing dialogue with different characters in response to a user's selection" is a device or method that generates responses and dialogue specific to the character selected by the user.

[0797] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with anime and game characters in real time. This system is realized by the roles of a server, terminals, and users, and their cooperation.

[0798] Server Features

[0799] The server collects character information, analyzes voice input, analyzes emotions, generates responses, and transmits response data.

[0800] 1. Collecting and Learning Character Information:

[0801] The server collects dialogue and voice data provided by character rights holders. This data is preprocessed using Python and LibROSA to remove noise. The preprocessed data is then trained by a generative AI model (e.g., GPT-3). This training process allows the system to analyze the character's unique dialogue and personality, resulting in natural-sounding dialogue.

[0802] 2. Receive and analyze user voice input:

[0803] The user's speech is captured by the device and sent to a server, which converts the speech data into text using a speech recognition system, such as Google Speech-to-Text.

[0804] 3. Emotion analysis:

[0805] The server then sends the converted text data to IBM Watson's emotion analysis engine to analyze the user's emotional state. Through this process, the user's emotions (such as joy, sadness, anger, etc.) are identified.

[0806] 4. Generate response:

[0807] Based on the results of the emotion analysis, the generative AI model generates an appropriate response. For example, if the analysis finds that the user is sad, the generative AI model generates comforting words such as "What happened? Thank you for telling me." An example of a prompt sentence could be, "If the user is sad, please generate comforting words."

[0808] 5. Sending response data:

[0809] The server sends the generated response text data to the terminal, which converts this data into voice data and outputs it to the user.

[0810] Device Features

[0811] The terminal is responsible for acquiring user voice, converting it into text data, communicating with the server, and outputting response data as voice.

[0812] 1. Capture voice input and convert it to text:

[0813] The device uses a microphone to capture the user's voice, which is then converted into text data using, for example, Google Speech-to-Text.

[0814] 2. Communication with the server:

[0815] The converted text data is sent to the server, and at the same time, a process is performed to wait for and receive response data from the server.

[0816] 3. Transcribing and playing back response data:

[0817] The device receives the text data returned from the server and converts it into voice data using a speech synthesis system such as Amazon Polly, which then responds to the user through the speaker.

[0818] User Actions

[0819] Users can select a character using a terminal application and begin interacting with that character. The figure's responses are adjusted according to the user's emotional state, allowing for more natural and rich interactions.

[0820] 1. Character selection and dialogue initiation:

[0821] Users select their favorite character through the terminal application, and then speak to the character to activate the dialogue system.

[0822] 2. Receiving responses and reflecting emotions:

[0823] You can receive voice responses from the figure and enjoy natural dialogue that reflects your own emotional state.

[0824] Specific examples

[0825] For example, if a user says, "Today was not a very fun day," the device captures the voice data and converts it into text using Google Speech-to-Text. The server analyzes the text data using IBM Watson's emotion analysis engine and determines that the user is sad. The generative AI model generates a response, "What happened? Thank you for talking," based on the prompt, "If the user is sad, please generate words of comfort." The device converts this response into speech using Amazon Polly, and the figure replies, "What happened? Thank you for talking."

[0826] Through this process, the system achieves realistic and natural dialogue between the user and the character. Appropriate responses generated based on the analysis results of the emotion engine further improve the quality of the dialogue.

[0827] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0828] Step 1: Gather and learn about the character

[0829] The server collects dialogue and voice data provided by character rights holders. The voice data is preprocessed and noise-removed using Python and LibROSA. The input is the provided character's dialogue and voice data, and the output is clean voice data with noise removed.

[0830] Specific operation: The server loads the collected audio files with LibROSA, performs noise filtering using FFT (Fast Fourier Transform), and generates a preprocessed dataset.

[0831] Step 2: Training the generative AI model

[0832] The server trains a generative AI model (e.g., GPT-3) based on the preprocessed data. The input is the preprocessed dataset, and the output is a model that has learned the character's unique lines and personality.

[0833] How it works: The server uses the dataset to train the AI ​​model, building a model that can generate responses that reflect the character's lines and personality traits.

[0834] Step 3: Getting voice input

[0835] The device captures the user's voice as they speak into the microphone. The input is the user's voice data, and the output is the captured raw audio file.

[0836] What it does: The device's microphone captures the user's speech and records it in real time.

[0837] Step 4: Speech to text

[0838] The device sends the captured voice data to a speech recognition system such as Google Speech-to-Text and converts it into text data. The input is the user's voice data, and the output is the corresponding text data.

[0839] Specific operation: The device sends the audio file to the cloud API and receives the returned text data.

[0840] Step 5: Sentiment analysis of text data

[0841] The server receives the text data sent from the device and performs emotion analysis using IBM Watson's emotion analysis engine. The input is the text data, and the output is the analyzed emotional state.

[0842] Specific operation: The server runs the text data through a sentiment analysis engine, resulting in the user's emotion (e.g., joy, sadness, anger).

[0843] Step 6: Generate a response

[0844] The server generates an appropriate response using a generative AI model based on the results of the sentiment analysis. The input is the sentiment analysis result and a prompt, and the output is a generated text response. An example of a prompt is "If the user is sad, please generate comforting words."

[0845] Specific operation: The server inputs a prompt to the generative AI model: "If the user is sad, please generate words of comfort." The model generates a response text such as "What happened? Thank you for telling me."

[0846] Step 7: Transcribing response data

[0847] The device sends the text data sent from the server to a speech synthesis system such as Amazon Polly and converts it into voice data. The input is the response text data, and the output is the synthesized voice data.

[0848] Specific operation: The device sends text data to the speech synthesis API and receives the generated audio file.

[0849] Step 8: Outputting audio data

[0850] The device outputs voice data to the user through a speaker. The input is synthesized voice data, and the output is the voice played from the speaker.

[0851] What happens: The device speaker plays the audio file and the user listens to it.

[0852] Through these processing steps, the system enables natural and emotionally responsive interactions between the user and the character.

[0853] (Application example 2)

[0854] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0855] In conventional communication systems, the dialogue between the user and the character is mechanical and one-way, rather than emotionally responsive. Therefore, there is a need to realize more natural and meaningful dialogue. Furthermore, if appropriate product suggestions can be made based on the user's emotional state, the customer experience can be improved.

[0856] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice input from the user, means for converting the voice input into text data, means for recognizing emotions based on the analyzed text data, and means for generating a response and making product suggestions based on the user's emotional state. This makes the dialogue between the user and the character natural and meaningful according to emotions, and further enables appropriate product suggestions based on the user's emotional state.

[0857] A "user" is a person who uses this system to interact with characters.

[0858] "Voice input" refers to the voice data that a user speaks to the system through a microphone.

[0859] "Text data" refers to character data converted by a voice recognition system based on voice input.

[0860] "Analysis" refers to the process of analyzing text data using an emotion recognition engine or similar to understand the user's emotions and intentions.

[0861] "Emotion recognition" is the process of identifying a user's emotional state through analysis.

[0862] "Generating an appropriate response" is the process of generating a reply to establish a dialogue based on the user's emotional state.

[0863] "Product suggestion" is the act of recommending appropriate products based on the user's emotional state and the content of their conversation.

[0864] "Voice data" refers to data that has been converted back into voice format from text data using a voice synthesis system.

[0865] "Output" refers to the act of transmitting the generated audio data to the user through a speaker.

[0866] A "system" is a collection of hardware and software necessary to realize interaction between a user and a character.

[0867] The system for implementing this invention utilizes emotion analysis data to provide appropriate responses and product recommendations in a virtual store where users can interact with anime and game characters. The detailed configuration of this system and the method for realizing it are described below.

[0868] System Configuration

[0869] 1. Server

[0870] Character information gathering and learning:

[0871] The server collects and preprocesses the dialogue and voice data provided by the character rights holders. After preprocessing such as noise removal, the generative AI model learns the character's unique dialogue and personality.

[0872] Receiving and parsing voice input:

[0873] Users input their voice into the virtual store, which is then received by the server, where it is converted into text data using a voice recognition system.

[0874] Sentiment analysis of text data:

[0875] The server sends the analyzed text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) to identify the user's emotion.

[0876] Generate appropriate responses and product suggestions:

[0877] Based on the results of the emotion analysis, a generative AI model (e.g., OpenAI GPT-3) is used to generate appropriate responses, and in addition, product suggestions are made based on the user's emotional state.

[0878] Sending response data:

[0879] The generated response and product proposal are transmitted to the terminal as text data.

[0880] 2. Terminal

[0881] Capture voice input and convert it to text:

[0882] Voice data is captured by the user's microphone and converted into text data using a voice recognition system (e.g., Google Speech-to-Text).

[0883] Sending the transformed data and receiving the response:

[0884] The converted text data is sent to the server, and response data from the server is waited for and received.

[0885] Transcribing and playing response data:

[0886] The received text data is sent to a speech synthesis system (e.g., Amazon Polly) and converted into voice data. The generated voice data is played back through the speaker to respond to the user.

[0887] 3. Users

[0888] Character selection and conversation initiation:

[0889] Users can use the virtual store application, select their favorite character, and start interacting with them. When they speak, the microphone captures the sound and sends it to the device.

[0890] Receiving responses and reflecting emotions:

[0891] The device receives voice feedback responses, which are tailored to the user's emotional state based on an emotion recognition engine.

[0892] Examples and prompts

[0893] For example, if a user says "I'm not feeling well today" in a virtual store, the following happens:

[0894] 1. The device captures the voice data, converts it into text data, and sends it to the server.

[0895] 2. The server performs speech recognition and emotion analysis and determines that the text data "I'm not feeling well" represents a sad emotion.

[0896] 3. The server's generated AI model generates data suggesting products to lift the mood (e.g., pleasant-smelling aromas) along with a response such as, "How's your day? Tell us if you'd like."

[0897] 4. The response data and product suggestions are sent to the terminal and converted into voice by a voice synthesis system.

[0898] 5. The user is prompted with the message, "How's your day going? Let me know if you'd like." along with suggestions for aroma products.

[0899] Prompt Sentence Examples

[0900] "You are a character in a virtual store. A user says, 'I'm feeling down today.' You encourage the user and suggest products that will lift their spirits."

[0901] In this way, this system provides natural and meaningful responses and product suggestions that correspond to the user's emotions through dialogue between the user and the character.

[0902] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0903] Step 1:

[0904] Users simply launch the virtual store application and select their favorite character, which then appears on the screen and can begin interacting with it.

[0905] Step 2:

[0906] The user speaks to the character, saying, "I'm not feeling well today." The device captures this voice data through the microphone and receives it as input.

[0907] Step 3:

[0908] The device uses a voice recognition system (e.g., Google Speech-to-Text) to convert the acquired voice data into text data. At this time, the device processes the voice recognition API and obtains the text data as output.

[0909] Step 4:

[0910] The terminal sends the obtained text data to the server. The text data is transferred as input to the server via the network, and the server receives it.

[0911] Step 5:

[0912] The server sends the received text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) for emotion analysis. The text data is passed as input to the analysis engine, and the analyzed user's emotional state is obtained as output.

[0913] Step 6:

[0914] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-3) based on the results of the emotion analysis. At this time, the prompt sentence is given as input to the generative AI model, and response text data that matches the user's emotion is obtained as output.

[0915] Step 7:

[0916] In addition to the generated responses, the server also generates product suggestions based on the user's emotional state. Using the emotion analysis results as input, the server uses a generative AI model to generate text data containing product suggestions as output.

[0917] Step 8:

[0918] The server transmits text data including the response and the product proposal to the terminal, and transmits the text data generated as input to the terminal as output via the network.

[0919] Step 9:

[0920] The device sends the received text data to a speech synthesis system (e.g., Amazon Polly) and converts it into voice data. Text data is passed as input to the speech synthesis API, and the generated voice data is obtained as output.

[0921] Step 10:

[0922] The generated voice data is played from the device's speaker to deliver a response to the user. Specifically, the device outputs the generated voice data to the speaker, and the user listens to it.

[0923] Step 11:

[0924] The user receives responses and suggestions from the character, and decides whether to continue the conversation or consider a product, thus creating a natural and emotionally relevant continuous interaction between the user and the character.

[0925] Through the above processing steps, the user can have a rich interaction experience with the character and can also receive product suggestions according to their emotional state.

[0926] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0927] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0928] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0929] [Third embodiment]

[0930] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0931] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0932] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0933] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0934] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0935] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0936] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0937] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0938] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0939] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0940] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0941] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0942] This invention relates to a communication system that allows users to have natural conversations with characters in anime and games in real time. The program processing of this system will be specifically described below.

[0943] server

[0944] 1. Collecting and Learning Character Information:

[0945] The server collects dialogue and voice data provided by character rights holders, and the collected data is pre-processed to remove noise and check grammar.

[0946] Based on the pre-processed data, the generative AI model learns the characters' unique lines and personalities, and the learning results are stored in a database.

[0947] 2. Generating and sending conversation data:

[0948] The server receives voice input from the user's device, which is converted into text data by a speech recognition system.

[0949] The converted text data is analyzed using natural language processing (NLP) and processed to understand the user's intent.

[0950] Based on the analysis results, an AI model is used to generate an appropriate response, which is generated as text data.

[0951] Finally, the generated text data is sent to the terminal.

[0952] Terminal

[0953] 1. Capture voice input and convert it to text:

[0954] The device receives the user's voice input and captures the voice data through the microphone, which is then sent to a voice recognition system and converted into text data.

[0955] 2. Sending a conversation request and receiving a response:

[0956] The converted text data is sent to the server and a response from the server is awaited.

[0957] The terminal receives the response data returned from the server.

[0958] 3. Transcribing and playing back response data:

[0959] The received text data is sent to a speech synthesis system and converted into voice data.

[0960] The generated audio data is played back through the figure's speaker so that the user can hear it.

[0961] user

[0962] 1. Character selection and conversation start:

[0963] Users use a terminal application to select their favorite character, and this information is stored on the terminal.

[0964] When you speak to the selected character, the figure will begin a conversation.

[0965] 2. Receiving the response:

[0966] The user receives a voice response from the figure, and repeats the process if they wish to continue the conversation.

[0967] Specific examples

[0968] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[0969] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[0970] The processing flow will be explained below.

[0971] server

[0972] Step 1:

[0973] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[0974] Step 2:

[0975] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[0976] Step 3:

[0977] The preprocessed data is analyzed and the speech is converted into text data using a speech recognition system.

[0978] Step 4:

[0979] The text data is used to further train the generative AI model, which learns dialogue based on information about the character's personality and behavior.

[0980] Step 5:

[0981] The learning results are stored in a database and can be accessed immediately upon user request.

[0982] Step 6:

[0983] The user sends voice input via the terminal to the server, which receives it and converts it into text data using a voice recognition system.

[0984] Step 7:

[0985] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intentions and emotions.

[0986] Step 8:

[0987] Based on the analysis results, a generative AI model is used to generate an appropriate response, which is then generated as text data.

[0988] Step 9:

[0989] The generated text data is sent to the terminal and a response is returned to the user.

[0990] Terminal

[0991] Step 1:

[0992] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[0993] Step 2:

[0994] The captured voice data is sent to a voice recognition system and converted into text data.

[0995] Step 3:

[0996] Sends the converted text data to the server and waits for a response from the server.

[0997] Step 4:

[0998] Receives response data from the server in text format.

[0999] Step 5:

[1000] The received text data is sent to a speech synthesis system and converted into voice data.

[1001] Step 6:

[1002] The generated audio data is played through the figure's speaker, and the user hears an audible response from the figure.

[1003] user

[1004] Step 1:

[1005] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[1006] Step 2:

[1007] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[1008] Step 3:

[1009] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[1010] Step 4:

[1011] The user receives a voice response from the figure and continues the conversation. By repeating the same steps, the user can continue to have natural conversations with the character.

[1012] Through these processing steps, the system enables realistic and natural interactions between users and characters. By having each entity work together, it is possible to improve the quality of the user experience and provide new business opportunities for character rights holders.

[1013] Example 1

[1014] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1015] In systems that allow users to have natural conversations with anime and game characters in real time, existing technologies have had difficulty adapting character-specific lines and personalities in real time. Furthermore, when changing characters according to user selection, the technology lacked the flexibility to provide different conversation patterns for each character.

[1016] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1017] In this invention, the server includes: [means for acquiring voice input from a user;] [means for converting the voice input into text data;] [means for analyzing the text data and generating an appropriate response;] [means for learning the character's unique lines and personality using a generative AI model;] [means for converting the generated response from text data into voice data; and [means for outputting the voice data to the user.] This allows the user to enjoy natural conversations in real time with the selected character.

[1018] "Means for obtaining voice input from a user" means devices and software for capturing the user's speech as digital voice data.

[1019] "Means for converting voice input into text data" refers to software and algorithms for converting captured voice data into text information using voice recognition technology.

[1020] "Means for analyzing text data and generating appropriate responses" refers to software and algorithms that analyze the converted text data using natural language processing technology, understand the user's intent, and generate appropriate text responses.

[1021] "Means for using a generative AI model to learn character-specific lines and personality traits" refers to a technology that uses provided character data to train a machine learning model to learn a character's specific speech patterns and personality traits.

[1022] The "means for converting the generated response from text data to speech data" is software and algorithms for converting the generated text data into natural-sounding speech data using text-to-speech (TTS) technology.

[1023] "Means for outputting audio data to a user" refers to equipment and software for playing the generated audio data through a speaker or other audio output device so that the user can hear it.

[1024] This invention relates to a communication system that allows users to have natural conversations with anime and game characters in real time. The system utilizes a generative AI model to convert the user's voice input into text data, analyze the text data, and generate appropriate responses.

[1025] server

[1026] 1. Collecting and Learning Character Information:

[1027] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed, which includes noise removal and grammar checks. Based on the preprocessed data, a generative AI model (e.g., a general-purpose machine learning model) learns the character's unique dialogue and personality. The learning results are stored in a database.

[1028] 2. Receiving voice input and converting it to text:

[1029] The server receives voice input from the user's device and converts the received voice data into text data using a voice recognition system (e.g., a general voice recognition API).

[1030] 3. Analyzing user intent:

[1031] The server then analyzes the converted text data using natural language processing (NLP) techniques, which provides the basis for understanding the user's intent and generating an appropriate response.

[1032] 4. Generate response data:

[1033] The server generates an appropriate response based on the parsed text data, using a generative AI model, and sends the generated text data to the device as a response.

[1034] Terminal

[1035] 1. Acquiring voice input:

[1036] The device captures the user's voice input through the microphone, and the captured voice data is temporarily stored in a buffer on the device.

[1037] 2. Send a request for text conversion:

[1038] The device sends the stored voice data to the server and makes a request to convert the voice to text.

[1039] 3. Receiving a response from the server:

[1040] The terminal receives the response data sent from the server, and this received data is temporarily stored in the terminal's memory.

[1041] 4. Speech synthesis and playback:

[1042] The device sends the received text data to a speech synthesis system (e.g., general speech synthesis software) and converts it into voice data, which is then played back through the figure's speaker.

[1043] user

[1044] 1. Character Selection:

[1045] Users select their favorite character using the device application, and this information is stored on the device and also sent to the server.

[1046] 2. Voice input:

[1047] Users can speak to the character they choose, for example, by saying, "How was your day?"

[1048] 3. Receiving the response:

[1049] The user hears a response from the figure, such as "Today was a fun day."

[1050] Specific examples

[1051] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[1052] Prompt Sentence Examples

[1053] "Design a system that allows a character figure to interact with the user in a natural way. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[1054] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[1055] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1056] Step 1:

[1057] The user selects a character. The user operates the application on the device and selects the character they like. This information is stored on the device and also sent to the server. The input is the user's selection, and the output is the selected character information.

[1058] Step 2:

[1059] The user provides voice input. The user speaks what they want to say to the character into the microphone. The input is the user's voice, and the output is saved as audio data on the device.

[1060] Step 3:

[1061] The device captures audio input. The device uses a microphone to capture the user's voice as digital audio data and stores it in a buffer. The input is the user's voice, and the output is digital audio data.

[1062] Step 4:

[1063] The device sends voice data to the server. The device sends saved voice data to the server and makes a voice recognition request. The input is voice data, and the output is voice data sent to the server.

[1064] Step 5:

[1065] The server converts the voice data into text data. The server converts the voice data into text information using a voice recognition system (for example, a general voice recognition API). The input is voice data, and the output is text data.

[1066] Step 6:

[1067] The server analyzes the text data. The server uses natural language processing (NLP) technology to analyze the text data and understand the user's intent. The input is the text data, and the output is the analysis result.

[1068] Step 7:

[1069] The server generates an appropriate response. The server uses a generative AI model to generate a response based on the analysis results. This response is generated as text data. The input is the analysis results, and the output is the generated text data.

[1070] Step 8:

[1071] The server sends the generated response to the terminal. The server sends the generated text data to the terminal. The input is the generated text data, and the output is the text data sent to the terminal.

[1072] Step 9:

[1073] The device converts text data into speech data. The device uses a speech synthesis system (e.g., common speech synthesis software) to convert text data into speech data. The input is text data and the output is speech data.

[1074] Step 10:

[1075] The terminal plays the generated audio data through the figure's speaker. The input is the audio data and the output is the played audio.

[1076] Example prompt sentence:

[1077] "Design a system that allows natural dialogue between a character figure and the user. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[1078] (Application example 1)

[1079] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1080] In virtual stores, the experience for users to obtain product information has not been intuitive enough, and effective natural dialogue has not been realized. Conventional systems lack the technology to enable natural dialogue between users and characters, resulting in a poor user experience. Therefore, there is a need for a method to enable users to efficiently obtain product information and improve satisfaction through more natural, real-time dialogue.

[1081] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1082] In this invention, the server includes: [means for acquiring voice input from a user]; [means for converting the voice input into text data]; [means for analyzing the text data and generating an appropriate response]; [means for converting the generated response from text data into voice data]; [means for outputting the voice data to the user]; and [means for a character in a virtual store to provide product information and improve the user's interaction experience.] This enables the user to obtain product information through natural interaction with the character in the virtual store.

[1083] "Capturing user voice input" is the process of capturing voice input data from a user and incorporating it into the system.

[1084] "Speech-to-text" is the process of converting captured voice data into text data.

[1085] "Text data analysis and response generation" is the process of analyzing the converted text data, understanding the user's intent, and generating an appropriate response.

[1086] "Conversion to voice data" refers to the process of converting the text data of the generated response into voice data.

[1087] "Outputting audio data" is the process of making the converted audio data audible to the user through a device such as a speaker.

[1088] "Learning character-specific lines and voice information" is the process of learning lines and voice characteristics specific to a particular character and using them to conduct natural conversations.

[1089] "Providing dialogue with different characters" is a process in which different characters are used depending on the user's selection, and the characters provide dialogue with the user.

[1090] "Providing product information in a virtual store" is the process by which a character in a virtual environment provides information about products to users.

[1091] "Improving the user's interaction experience" is a process aimed at qualitatively improving the experience that users have through natural interactions with characters.

[1092] This program is a system that allows users to obtain product information through natural dialogue with characters in a virtual store and assists the purchasing process. The system is composed of a server, a terminal, and user operations.

[1093] server

[1094] The server has the following responsibilities:

[1095] 1. Collecting and Learning Character Information:

[1096] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed to remove noise and check grammar. Based on this preprocessed data, the generative AI model learns the character's unique dialogue and personality. The learning results are stored in a database.

[1097] 2. Generating and sending conversation data:

[1098] The server receives voice input from the device and converts it into text using a speech recognition system. The converted text is then analyzed using natural language processing (NLP) to understand the user's intent and generate an appropriate response, which is then sent back to the device.

[1099] Terminal

[1100] The terminal has the following responsibilities:

[1101] 1. Capture voice input and convert it to text:

[1102] It takes the user's voice input and captures it through a microphone, which sends the voice data to a speech recognition system and converts it into text data.

[1103] 2. Sending a conversation request and receiving a response:

[1104] The converted text data is sent to the server, and a response is waited for from the server. The terminal again receives the response data sent back from the server.

[1105] 3. Transcribing and playing back response data:

[1106] The received text data is converted into voice data using a speech synthesis system, and the generated voice data is played back through the device's speaker, allowing the user to listen to it.

[1107] user

[1108] The user operates the following functions:

[1109] 1. Character selection and conversation start:

[1110] Users use a terminal application to select their favorite character, which is stored on the terminal, and can then start a conversation by speaking to the selected character.

[1111] 2. Receiving the response:

[1112] The user receives a voice response from the character, and if they wish to continue the conversation, they repeat the same process.

[1113] Hardware and software used

[1114] Hardware: smartphone, microphone, speaker

[1115] Software: Python, SpeechRecognition library, Transformers library, gTTS (Google Text-to-Speech)

[1116] Specific examples

[1117] When a user asks the figure, "What products do you recommend?", the following process takes place:

[1118] The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "The recommended product is the latest smartphone," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figure replies, "The recommended product is the latest smartphone." In this way, the user enjoys a realistic dialogue with the character.

[1119] Prompt Sentence Examples

[1120] 1. "What products do you recommend?"

[1121] 2. "Please tell me the features of this product."

[1122] 3. "Do you have it in stock?"

[1123] 4. "When is the next sale?"

[1124] This allows the system to enable users to obtain product information through natural interactions with characters in the virtual store, improving the user experience.

[1125] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1126] Step 1:

[1127] The user speaks into the device's microphone, and the user's voice input is captured through the device's microphone.

[1128] Input: User's voice

[1129] Output: Audio data

[1130] Step 2:

[1131] The device sends the captured voice data to a speech recognition system that converts it into text using the SpeechRecognition library.

[1132] Input: Audio data

[1133] Output: Text data

[1134] Step 3:

[1135] The text data is transmitted from the terminal to the server.

[1136] Input: Text data

[1137] Output: Text data sent to the server

[1138] Step 4:

[1139] The server analyzes the received text data using natural language processing (NLP). Specifically, a generative AI model using the Transformers library analyzes the text data and understands the user's intent.

[1140] Input: Text data sent to the server

[1141] Output: Analysis results

[1142] Step 5:

[1143] The server generates an appropriate response based on the analysis results, using a generative AI model to generate the best response to the user's question or request.

[1144] Input: Analysis results

[1145] Output: Response text data

[1146] Step 6:

[1147] The generated response text data is transmitted from the server to the terminal.

[1148] Input: Response text data

[1149] Output: Response text data sent to the terminal

[1150] Step 7:

[1151] The device sends the received response text data to a speech synthesis system, which converts it into voice data. The speech synthesis system uses gTTS (Google Text-to-Speech).

[1152] Input: Response text data sent to the terminal

[1153] Output: Response audio data

[1154] Step 8:

[1155] The generated audio data is played back through the device's speaker, allowing the user to hear it.

[1156] Input: Response audio data

[1157] Output: The response the user hears

[1158] To summarize the specific operations of each step, the user speaks a question into the device, the speech is converted into text and sent to the server, the server analyzes the text data, generates an appropriate response and sends it back to the device, and the device converts the received text response into speech and plays it back to the user, achieving natural dialogue in real time.

[1159] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1160] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with characters in anime or games in real time. The program processing of this system will be specifically described below.

[1161] server

[1162] 1. Collecting and Learning Character Information:

[1163] The server collects dialogue and voice data provided by character rights holders. This data is pre-processed to remove unwanted noise. Based on the pre-processed data, the generative AI model learns the character's unique dialogue and personality.

[1164] 2. Receiving and parsing voice input from the user:

[1165] When a user makes a voice input, the data is sent via the terminal to the server, which receives the voice data and converts it into text data using a voice recognition system.

[1166] 3. Sentiment Analysis of Text Data:

[1167] The server sends the converted text data to the emotion engine, which analyzes the user's emotions and determines their emotional state from the tone and content of their voice.

[1168] 4. Generating an appropriate response:

[1169] Based on the results of the emotion analysis, the server uses a generative AI model to generate an appropriate response. For example, if the user looks sad, the character will respond with comforting words.

[1170] 5. Sending response data:

[1171] The generated response data is sent to the terminal and a reply is sent to the user.

[1172] Terminal

[1173] 1. Capture voice input and convert it to text:

[1174] The device uses a microphone to capture the user's voice, which is then sent to a voice recognition system and converted into text.

[1175] 2. Sending the converted data and receiving the response:

[1176] The terminal transmits the converted text data to the server, waits for response data from the server, and receives the text data returned from the server.

[1177] 3. Transcribing and playing back response data:

[1178] The device sends the received text data to a speech synthesis system, converts it into voice data, and plays the generated voice data through the figure's speaker to respond to the user.

[1179] user

[1180] 1. Character selection and conversation start:

[1181] Users use the terminal application to select their favorite character, and when they speak to the selected character, the figure captures their voice.

[1182] 2. Receiving responses and reflecting emotions:

[1183] The figure responds with a voice response that is based on the user's emotional state, and the emotion engine adjusts the response to take the user's state into account, allowing for more natural and meaningful interactions.

[1184] Specific examples

[1185] For example, if a user says, "Today was not a very fun day," the following process occurs: The device captures the voice data, converts it into text data, and sends it to the server. The server analyzes the text data with an emotion engine and determines that the user is sad. The generative AI model generates comforting words such as "What happened? Thanks for telling me," and sends the text data back to the device. The device converts this data into speech, and the figure replies, "What happened? Thanks for telling me." In this way, the user can continue the dialogue with the character in a natural way.

[1186] Through these processing steps, the system achieves realistic and natural dialogue between the user and the character. The incorporation of an emotion engine generates appropriate responses based on the user's emotions, further improving the quality of the user experience.

[1187] The processing flow will be explained below.

[1188] server

[1189] Step 1:

[1190] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[1191] Step 2:

[1192] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[1193] Step 3:

[1194] Analyze the pre-processed data and convert the speech into text data. Use a speech recognition system to efficiently convert the collected speech information into text.

[1195] Step 4:

[1196] The text data is used to further train the generative AI model, which then incorporates the character's unique lines and personality traits and updates its parameters.

[1197] Step 5:

[1198] The learning results are stored in a database to enable quick responses to user requests.

[1199] Step 6:

[1200] When a user sends a voice input to the server via the terminal, the server receives it and converts the voice data into text data using a voice recognition system.

[1201] Step 7:

[1202] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intent.

[1203] Step 8:

[1204] Analyze user sentiment from text data using an emotion engine, which identifies emotions from the user's phrasing and tone of voice.

[1205] Step 9:

[1206] Based on the analysis results, an appropriate response is generated using a generative AI model, which is then structured as text data.

[1207] Step 10:

[1208] The generated text data is sent to the terminal and a reply is given to the user, which is tailored to the user's emotions.

[1209] Terminal

[1210] Step 1:

[1211] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[1212] Step 2:

[1213] The captured voice data is sent to a voice recognition system and converted into text data using voice recognition software.

[1214] Step 3:

[1215] The converted text data is sent to the server and a response is awaited from the server, a necessary step to prepare the server for processing the data.

[1216] Step 4:

[1217] Receives the response data received from the server in text format. This data was generated by the server based on the user's request.

[1218] Step 5:

[1219] The received text data is sent to a speech synthesis system and converted into voice data. Through the speech synthesis process, the text data is turned into voice.

[1220] Step 6:

[1221] The generated voice data is played through the figure's speaker, and the figure uses this voice data to respond to the user.

[1222] user

[1223] Step 1:

[1224] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[1225] Step 2:

[1226] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[1227] Step 3:

[1228] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[1229] Step 4:

[1230] The user can continue the conversation by receiving voice responses from the figure. The responses correspond to the user's emotions, allowing for natural dialogue.

[1231] Examples:

[1232] For example, if a user says to a figurine, "You seem tired and down lately," the following process takes place: The device captures the voice data and converts it into text data using a voice recognition system. This text data is sent to the server, where an emotion engine analyzes the emotion "tired." Based on the analysis results, the generative AI model generates a response such as "Are you OK? You should get some rest," and sends it back to the device as text data. The device converts this data into voice, and the figurine responds, "Are you OK? You should get some rest." In this way, the user can continue to have a natural conversation with the character.

[1233] Through these specific processing steps, the system achieves realistic and natural dialogue between users and characters. The built-in emotion engine generates appropriate responses based on the user's emotions, further improving the user experience and providing new business opportunities for character rights holders.

[1234] Example 2

[1235] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1236] Conventional communication systems are unable to generate appropriate responses based on the user's emotional state, resulting in conversations that tend to be one-sided and mechanical. Furthermore, they lack the means to learn character-specific lines and personalities, resulting in a poor user experience. There is a need for a communication system that can solve these issues and realize more natural and emotionally sensitive conversations.

[1237] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1238] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data with an emotion analysis engine and identifying the user's emotional state, means for generating an appropriate response using a generative AI model based on the results of the emotion analysis, means for converting the generated response from text data into voice data, and means for outputting the voice data to the user, thereby enabling natural and meaningful dialogue that is in tune with the user's emotional state.

[1239] "Means for obtaining voice input from a user" refers to a device or method for collecting voice data uttered by a user using a microphone or voice input device.

[1240] The "means for converting speech input into text data" refers to a device or method that uses a speech recognition system to convert acquired speech data into text information.

[1241] "Means for analyzing text data using an emotion analysis engine and identifying a user's emotional state" refers to a device or method that uses text data as input and analyzes and identifies a user's emotional state (e.g., joy, sadness, anger, etc.).

[1242] "Means for generating an appropriate response using a generative AI model based on the results of emotion analysis" refers to a device or method that uses a generative AI model to create a natural and appropriate response, taking into account the user's emotional state obtained through emotion analysis.

[1243] The "means for converting the generated response from text data to voice data" refers to a device or method for converting a response expressed in text data into voice data using a voice synthesis system.

[1244] "Means for outputting audio data to a user" means a device or method for communicating the generated audio data to a user through a speaker or other audio output device.

[1245] "Means for learning character-specific lines and personality information" refers to the process of using a machine learning model to learn given lines and personality information to reflect the character's personality and speech.

[1246] A "means for providing dialogue with different characters in response to a user's selection" is a device or method that generates responses and dialogue specific to the character selected by the user.

[1247] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with anime and game characters in real time. This system is realized by the roles of a server, terminals, and users, and their cooperation.

[1248] Server Features

[1249] The server collects character information, analyzes voice input, analyzes emotions, generates responses, and transmits response data.

[1250] 1. Collecting and Learning Character Information:

[1251] The server collects dialogue and voice data provided by character rights holders. This data is preprocessed using Python and LibROSA to remove noise. The preprocessed data is then trained by a generative AI model (e.g., GPT-3). This training process allows the system to analyze the character's unique dialogue and personality, resulting in natural-sounding dialogue.

[1252] 2. Receive and analyze user voice input:

[1253] The user's speech is captured by the device and sent to a server, which converts the speech data into text using a speech recognition system, such as Google Speech-to-Text.

[1254] 3. Emotion analysis:

[1255] The server then sends the converted text data to IBM Watson's emotion analysis engine to analyze the user's emotional state. Through this process, the user's emotions (such as joy, sadness, anger, etc.) are identified.

[1256] 4. Generate response:

[1257] Based on the results of the emotion analysis, the generative AI model generates an appropriate response. For example, if the analysis finds that the user is sad, the generative AI model generates comforting words such as "What happened? Thank you for telling me." An example of a prompt sentence could be, "If the user is sad, please generate comforting words."

[1258] 5. Sending response data:

[1259] The server sends the generated response text data to the terminal, which converts this data into voice data and outputs it to the user.

[1260] Device Features

[1261] The terminal is responsible for acquiring user voice, converting it into text data, communicating with the server, and outputting response data as voice.

[1262] 1. Capture voice input and convert it to text:

[1263] The device uses a microphone to capture the user's voice, which is then converted into text data using, for example, Google Speech-to-Text.

[1264] 2. Communication with the server:

[1265] The converted text data is sent to the server, and at the same time, a process is performed to wait for and receive response data from the server.

[1266] 3. Transcribing and playing back response data:

[1267] The device receives the text data returned from the server and converts it into voice data using a speech synthesis system such as Amazon Polly, which then responds to the user through the speaker.

[1268] User Actions

[1269] Users can select a character using a terminal application and begin interacting with that character. The figure's responses are adjusted according to the user's emotional state, allowing for more natural and rich interactions.

[1270] 1. Character selection and dialogue initiation:

[1271] Users select their favorite character through the terminal application, and then speak to the character to activate the dialogue system.

[1272] 2. Receiving responses and reflecting emotions:

[1273] You can receive voice responses from the figure and enjoy natural dialogue that reflects your own emotional state.

[1274] Specific examples

[1275] For example, if a user says, "Today was not a very fun day," the device captures the voice data and converts it into text using Google Speech-to-Text. The server analyzes the text data using IBM Watson's emotion analysis engine and determines that the user is sad. The generative AI model generates a response, "What happened? Thank you for talking," based on the prompt, "If the user is sad, please generate words of comfort." The device converts this response into speech using Amazon Polly, and the figure replies, "What happened? Thank you for talking."

[1276] Through this process, the system achieves realistic and natural dialogue between the user and the character. Appropriate responses generated based on the analysis results of the emotion engine further improve the quality of the dialogue.

[1277] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1278] Step 1: Gather and learn about the character

[1279] The server collects dialogue and voice data provided by character rights holders. The voice data is preprocessed and noise-removed using Python and LibROSA. The input is the provided character's dialogue and voice data, and the output is clean voice data with noise removed.

[1280] Specific operation: The server loads the collected audio files with LibROSA, performs noise filtering using FFT (Fast Fourier Transform), and generates a preprocessed dataset.

[1281] Step 2: Training the generative AI model

[1282] The server trains a generative AI model (e.g., GPT-3) based on the preprocessed data. The input is the preprocessed dataset, and the output is a model that has learned the character's unique lines and personality.

[1283] How it works: The server uses the dataset to train the AI ​​model, building a model that can generate responses that reflect the character's lines and personality traits.

[1284] Step 3: Getting voice input

[1285] The device captures the user's voice as they speak into the microphone. The input is the user's voice data, and the output is the captured raw audio file.

[1286] What it does: The device's microphone captures the user's speech and records it in real time.

[1287] Step 4: Speech to text

[1288] The device sends the captured voice data to a speech recognition system such as Google Speech-to-Text and converts it into text data. The input is the user's voice data, and the output is the corresponding text data.

[1289] Specific operation: The device sends the audio file to the cloud API and receives the returned text data.

[1290] Step 5: Sentiment analysis of text data

[1291] The server receives the text data sent from the device and performs emotion analysis using IBM Watson's emotion analysis engine. The input is the text data, and the output is the analyzed emotional state.

[1292] Specific operation: The server runs the text data through a sentiment analysis engine, resulting in the user's emotion (e.g., joy, sadness, anger).

[1293] Step 6: Generate a response

[1294] The server generates an appropriate response using a generative AI model based on the results of the sentiment analysis. The input is the sentiment analysis result and a prompt, and the output is a generated text response. An example of a prompt is "If the user is sad, please generate comforting words."

[1295] Specific operation: The server inputs a prompt to the generative AI model: "If the user is sad, please generate words of comfort." The model generates a response text such as "What happened? Thank you for telling me."

[1296] Step 7: Transcribing response data

[1297] The device sends the text data sent from the server to a speech synthesis system such as Amazon Polly and converts it into voice data. The input is the response text data, and the output is the synthesized voice data.

[1298] Specific operation: The device sends text data to the speech synthesis API and receives the generated audio file.

[1299] Step 8: Outputting audio data

[1300] The device outputs voice data to the user through a speaker. The input is synthesized voice data, and the output is the voice played from the speaker.

[1301] What happens: The device speaker plays the audio file and the user listens to it.

[1302] Through these processing steps, the system enables natural and emotionally responsive interactions between the user and the character.

[1303] (Application example 2)

[1304] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1305] In conventional communication systems, the dialogue between the user and the character is mechanical and one-way, rather than emotionally responsive. Therefore, there is a need to realize more natural and meaningful dialogue. Furthermore, if appropriate product suggestions can be made based on the user's emotional state, the customer experience can be improved.

[1306] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice input from the user, means for converting the voice input into text data, means for recognizing emotions based on the analyzed text data, and means for generating a response and making product suggestions based on the user's emotional state. This makes the dialogue between the user and the character natural and meaningful according to emotions, and further enables appropriate product suggestions based on the user's emotional state.

[1307] A "user" is a person who uses this system to interact with characters.

[1308] "Voice input" refers to the voice data that a user speaks to the system through a microphone.

[1309] "Text data" refers to character data converted by a voice recognition system based on voice input.

[1310] "Analysis" refers to the process of analyzing text data using an emotion recognition engine or similar to understand the user's emotions and intentions.

[1311] "Emotion recognition" is the process of identifying a user's emotional state through analysis.

[1312] "Generating an appropriate response" is the process of generating a reply to establish a dialogue based on the user's emotional state.

[1313] "Product suggestion" is the act of recommending appropriate products based on the user's emotional state and the content of their conversation.

[1314] "Voice data" refers to data that has been converted back into voice format from text data using a voice synthesis system.

[1315] "Output" refers to the act of transmitting the generated audio data to the user through a speaker.

[1316] A "system" is a collection of hardware and software necessary to realize interaction between a user and a character.

[1317] The system for implementing this invention utilizes emotion analysis data to provide appropriate responses and product recommendations in a virtual store where users can interact with anime and game characters. The detailed configuration of this system and the method for realizing it are described below.

[1318] System Configuration

[1319] 1. Server

[1320] Character information gathering and learning:

[1321] The server collects and preprocesses the dialogue and voice data provided by the character rights holders. After preprocessing such as noise removal, the generative AI model learns the character's unique dialogue and personality.

[1322] Receiving and parsing voice input:

[1323] Users input their voice into the virtual store, which is then received by the server, where the voice data is converted into text data using a voice recognition system.

[1324] Sentiment analysis of text data:

[1325] The server sends the analyzed text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) to identify the user's emotion.

[1326] Generate appropriate responses and product suggestions:

[1327] Based on the results of the emotion analysis, a generative AI model (e.g., OpenAI GPT-3) is used to generate appropriate responses, and in addition, product suggestions are made based on the user's emotional state.

[1328] Sending response data:

[1329] The generated response and product proposal are transmitted to the terminal as text data.

[1330] 2. Terminal

[1331] Capture voice input and convert it to text:

[1332] Voice data is captured by the user's microphone and converted into text data using a voice recognition system (e.g., Google Speech-to-Text).

[1333] Sending the transformed data and receiving the response:

[1334] The converted text data is sent to the server, and response data from the server is waited for and received.

[1335] Transcribing and playing response data:

[1336] The received text data is sent to a speech synthesis system (e.g., Amazon Polly) and converted into voice data. The generated voice data is played back through the speaker to respond to the user.

[1337] 3. Users

[1338] Character selection and conversation initiation:

[1339] Users can use the virtual store application, select their favorite character, and start interacting with them. When they speak, the microphone captures the sound and sends it to the device.

[1340] Receiving responses and reflecting emotions:

[1341] The device receives voice feedback responses, which are tailored to the user's emotional state based on an emotion recognition engine.

[1342] Examples and prompts

[1343] For example, if a user says "I'm not feeling well today" in a virtual store, the following happens:

[1344] 1. The device captures the voice data, converts it into text data, and sends it to the server.

[1345] 2. The server performs speech recognition and emotion analysis and determines that the text data "I'm not feeling well" represents a sad emotion.

[1346] 3. The server's generated AI model generates data suggesting products to lift the mood (e.g., pleasant-smelling aromas) along with a response such as, "How's your day? Tell us if you'd like."

[1347] 4. The response data and product suggestions are sent to the terminal and converted into voice by a voice synthesis system.

[1348] 5. The user is prompted with the message, "How's your day going? Let me know if you'd like." along with suggestions for aroma products.

[1349] Prompt Sentence Examples

[1350] "You are a character in a virtual store. A user says, 'I'm feeling down today.' You encourage the user and suggest products that will lift their spirits."

[1351] In this way, this system provides natural and meaningful responses and product suggestions that correspond to the user's emotions through dialogue between the user and the character.

[1352] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1353] Step 1:

[1354] Users simply launch the virtual store application and select their favorite character, which then appears on the screen and can begin interacting with it.

[1355] Step 2:

[1356] The user speaks to the character, saying, "I'm not feeling well today." The device captures this voice data through the microphone and receives it as input.

[1357] Step 3:

[1358] The device uses a voice recognition system (e.g., Google Speech-to-Text) to convert the acquired voice data into text data. At this time, the device processes the voice recognition API and obtains the text data as output.

[1359] Step 4:

[1360] The terminal sends the obtained text data to the server. The text data is transferred as input to the server via the network, and the server receives it.

[1361] Step 5:

[1362] The server sends the received text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) for emotion analysis. The text data is passed as input to the analysis engine, and the analyzed user's emotional state is obtained as output.

[1363] Step 6:

[1364] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-3) based on the results of the emotion analysis. At this time, the prompt sentence is given as input to the generative AI model, and response text data that matches the user's emotion is obtained as output.

[1365] Step 7:

[1366] In addition to the generated responses, the server also generates product suggestions based on the user's emotional state. Using the emotion analysis results as input, the server uses a generative AI model to generate text data containing product suggestions as output.

[1367] Step 8:

[1368] The server transmits text data including the response and the product proposal to the terminal, and transmits the text data generated as input to the terminal as output via the network.

[1369] Step 9:

[1370] The device sends the received text data to a speech synthesis system (e.g., Amazon Polly) and converts it into voice data. Text data is passed as input to the speech synthesis API, and the generated voice data is obtained as output.

[1371] Step 10:

[1372] The generated voice data is played from the device's speaker to deliver a response to the user. Specifically, the device outputs the generated voice data to the speaker, and the user listens to it.

[1373] Step 11:

[1374] The user receives responses and suggestions from the character, and decides whether to continue the conversation or consider a product, thus creating a natural and emotionally relevant continuous interaction between the user and the character.

[1375] Through the above processing steps, the user can have a rich interaction experience with the character and can also receive product suggestions according to their emotional state.

[1376] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1377] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1378] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1379] [Fourth embodiment]

[1380] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1381] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1382] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1383] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1384] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1385] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1386] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1387] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1388] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1389] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1390] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1391] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1392] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1393] This invention relates to a communication system that allows users to have natural conversations with characters in anime and games in real time. The program processing of this system will be specifically described below.

[1394] server

[1395] 1. Collecting and Learning Character Information:

[1396] The server collects dialogue and voice data provided by character rights holders, and the collected data is pre-processed to remove noise and check grammar.

[1397] Based on the pre-processed data, the generative AI model learns the characters' unique lines and personalities, and the learning results are stored in a database.

[1398] 2. Generating and sending conversation data:

[1399] The server receives voice input from the user's device, which is converted into text data by a speech recognition system.

[1400] The converted text data is analyzed using natural language processing (NLP) and processed to understand the user's intent.

[1401] Based on the analysis results, an AI model is used to generate an appropriate response, which is generated as text data.

[1402] Finally, the generated text data is sent to the terminal.

[1403] Terminal

[1404] 1. Capture voice input and convert it to text:

[1405] The device receives the user's voice input and captures the voice data through the microphone, which is then sent to a voice recognition system and converted into text data.

[1406] 2. Sending a conversation request and receiving a response:

[1407] The converted text data is sent to the server and a response from the server is awaited.

[1408] The terminal receives the response data returned from the server.

[1409] 3. Transcribing and playing back response data:

[1410] The received text data is sent to a speech synthesis system and converted into voice data.

[1411] The generated audio data is played back through the figure's speaker so that the user can hear it.

[1412] user

[1413] 1. Character selection and conversation start:

[1414] Users use a terminal application to select their favorite character, and this information is stored on the terminal.

[1415] When you speak to the selected character, the figure will begin a conversation.

[1416] 2. Receiving the response:

[1417] The user receives a voice response from the figure, and repeats the process if they wish to continue the conversation.

[1418] Specific examples

[1419] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[1420] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[1421] The processing flow will be explained below.

[1422] server

[1423] Step 1:

[1424] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[1425] Step 2:

[1426] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[1427] Step 3:

[1428] The preprocessed data is analyzed and the speech is converted into text data using a speech recognition system.

[1429] Step 4:

[1430] The text data is used to further train the generative AI model, which learns dialogue based on information about the character's personality and behavior.

[1431] Step 5:

[1432] The learning results are stored in a database and can be accessed immediately upon user request.

[1433] Step 6:

[1434] The user sends voice input via the terminal to the server, which receives it and converts it into text data using a voice recognition system.

[1435] Step 7:

[1436] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intentions and emotions.

[1437] Step 8:

[1438] Based on the analysis results, a generative AI model is used to generate an appropriate response, which is then generated as text data.

[1439] Step 9:

[1440] The generated text data is sent to the terminal and a response is returned to the user.

[1441] Terminal

[1442] Step 1:

[1443] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[1444] Step 2:

[1445] The captured voice data is sent to a voice recognition system and converted into text data.

[1446] Step 3:

[1447] Sends the converted text data to the server and waits for a response from the server.

[1448] Step 4:

[1449] Receives response data from the server in text format.

[1450] Step 5:

[1451] The received text data is sent to a speech synthesis system and converted into voice data.

[1452] Step 6:

[1453] The generated audio data is played through the figure's speaker, and the user hears an audible response from the figure.

[1454] user

[1455] Step 1:

[1456] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[1457] Step 2:

[1458] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[1459] Step 3:

[1460] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[1461] Step 4:

[1462] The user receives a voice response from the figure and continues the conversation. By repeating the same steps, the user can continue to have natural conversations with the character.

[1463] Through these processing steps, the system enables realistic and natural interactions between users and characters. By having each entity work together, it is possible to improve the quality of the user experience and provide new business opportunities for character rights holders.

[1464] Example 1

[1465] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1466] In systems that allow users to have natural conversations with anime and game characters in real time, existing technologies have had difficulty adapting character-specific lines and personalities in real time. Furthermore, when changing characters according to user selection, the technology lacked the flexibility to provide different conversation patterns for each character.

[1467] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1468] In this invention, the server includes: [means for acquiring voice input from a user;] [means for converting the voice input into text data;] [means for analyzing the text data and generating an appropriate response;] [means for learning the character's unique lines and personality using a generative AI model;] [means for converting the generated response from text data into voice data; and [means for outputting the voice data to the user.] This allows the user to enjoy natural conversations in real time with the selected character.

[1469] "Means for obtaining voice input from a user" means devices and software for capturing the user's speech as digital voice data.

[1470] "Means for converting voice input into text data" refers to software and algorithms for converting captured voice data into text information using voice recognition technology.

[1471] "Means for analyzing text data and generating appropriate responses" refers to software and algorithms that analyze the converted text data using natural language processing technology, understand the user's intent, and generate appropriate text responses.

[1472] "Means for using a generative AI model to learn character-specific lines and personality traits" refers to a technology that uses provided character data to train a machine learning model to learn a character's specific speech patterns and personality traits.

[1473] The "means for converting the generated response from text data to speech data" is software and algorithms for converting the generated text data into natural-sounding speech data using text-to-speech (TTS) technology.

[1474] "Means for outputting audio data to a user" refers to equipment and software for playing the generated audio data through a speaker or other audio output device so that the user can hear it.

[1475] This invention relates to a communication system that allows users to have natural conversations with anime and game characters in real time. The system utilizes a generative AI model to convert the user's voice input into text data, analyze the text data, and generate appropriate responses.

[1476] server

[1477] 1. Collecting and Learning Character Information:

[1478] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed, which includes noise removal and grammar checks. Based on the preprocessed data, a generative AI model (e.g., a general-purpose machine learning model) learns the character's unique dialogue and personality. The learning results are stored in a database.

[1479] 2. Receiving voice input and converting it to text:

[1480] The server receives voice input from the user's device and converts the received voice data into text data using a voice recognition system (e.g., a general voice recognition API).

[1481] 3. Analyzing user intent:

[1482] The server then analyzes the converted text data using natural language processing (NLP) techniques, which provides the basis for understanding the user's intent and generating an appropriate response.

[1483] 4. Generate response data:

[1484] The server generates an appropriate response based on the parsed text data, using a generative AI model, and sends the generated text data to the device as a response.

[1485] Terminal

[1486] 1. Acquiring voice input:

[1487] The device captures the user's voice input through the microphone, and the captured voice data is temporarily stored in a buffer on the device.

[1488] 2. Send a request for text conversion:

[1489] The device sends the stored voice data to the server and makes a request to convert the voice to text.

[1490] 3. Receiving a response from the server:

[1491] The terminal receives the response data sent from the server, and this received data is temporarily stored in the terminal's memory.

[1492] 4. Speech synthesis and playback:

[1493] The device sends the received text data to a speech synthesis system (e.g., general speech synthesis software) and converts it into voice data, which is then played back through the figure's speaker.

[1494] user

[1495] 1. Character Selection:

[1496] Users select their favorite character using the device application, and this information is stored on the device and also sent to the server.

[1497] 2. Voice input:

[1498] Users can speak to the character they choose, for example, by saying, "How was your day?"

[1499] 3. Receiving the response:

[1500] The user hears a response from the figure, such as "Today was a fun day."

[1501] Specific examples

[1502] For example, if a user asks a figurine, "How was your day today?", the following process takes place: The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "It was a fun day," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figurine responds, "It was a fun day today." In this way, users can enjoy realistic dialogue with the character.

[1503] Prompt Sentence Examples

[1504] "Design a system that allows a character figure to interact with the user in a natural way. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[1505] As described above, this system provides a means for users to communicate with characters in a natural and real-time manner, improving the user experience and creating new business opportunities for character rights holders.

[1506] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1507] Step 1:

[1508] The user selects a character. The user operates the application on the device and selects the character they like. This information is stored on the device and also sent to the server. The input is the user's selection, and the output is the selected character information.

[1509] Step 2:

[1510] The user provides voice input. The user speaks what they want to say to the character into the microphone. The input is the user's voice, and the output is saved as audio data on the device.

[1511] Step 3:

[1512] The device captures audio input. The device uses a microphone to capture the user's voice as digital audio data and stores it in a buffer. The input is the user's voice, and the output is digital audio data.

[1513] Step 4:

[1514] The device sends voice data to the server. The device sends saved voice data to the server and makes a voice recognition request. The input is voice data, and the output is voice data sent to the server.

[1515] Step 5:

[1516] The server converts the voice data into text data. The server converts the voice data into text information using a voice recognition system (for example, a general voice recognition API). The input is voice data, and the output is text data.

[1517] Step 6:

[1518] The server analyzes the text data. The server uses natural language processing (NLP) technology to analyze the text data and understand the user's intent. The input is the text data, and the output is the analysis result.

[1519] Step 7:

[1520] The server generates an appropriate response. The server uses a generative AI model to generate a response based on the analysis results. This response is generated as text data. The input is the analysis results, and the output is the generated text data.

[1521] Step 8:

[1522] The server sends the generated response to the terminal. The server sends the generated text data to the terminal. The input is the generated text data, and the output is the text data sent to the terminal.

[1523] Step 9:

[1524] The device converts text data into speech data. The device uses a speech synthesis system (e.g., common speech synthesis software) to convert text data into speech data. The input is text data and the output is speech data.

[1525] Step 10:

[1526] The terminal plays the generated audio data through the figure's speaker. The input is the audio data and the output is the played audio.

[1527] Example prompt sentence:

[1528] "Design a system that allows natural dialogue between a character figure and the user. If the user asks, 'How was your day?' please explain in detail how this would be handled."

[1529] (Application example 1)

[1530] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1531] In virtual stores, the experience for users to obtain product information has not been intuitive enough, and effective natural dialogue has not been realized. Conventional systems lack the technology to enable natural dialogue between users and characters, resulting in a poor user experience. Therefore, there is a need for a method to enable users to efficiently obtain product information and improve satisfaction through more natural, real-time dialogue.

[1532] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1533] In this invention, the server includes: [means for acquiring voice input from a user]; [means for converting the voice input into text data]; [means for analyzing the text data and generating an appropriate response]; [means for converting the generated response from text data into voice data]; [means for outputting the voice data to the user]; and [means for a character in a virtual store to provide product information and improve the user's interaction experience.] This enables the user to obtain product information through natural interaction with the character in the virtual store.

[1534] "Capturing user voice input" is the process of capturing voice input data from a user and incorporating it into the system.

[1535] "Speech-to-text" is the process of converting captured voice data into text data.

[1536] "Text data analysis and response generation" is the process of analyzing the converted text data, understanding the user's intent, and generating an appropriate response.

[1537] "Conversion to voice data" refers to the process of converting the text data of the generated response into voice data.

[1538] "Outputting audio data" is the process of making the converted audio data audible to the user through a device such as a speaker.

[1539] "Learning character-specific lines and voice information" is the process of learning lines and voice characteristics specific to a particular character and using them to conduct natural conversations.

[1540] "Providing dialogue with different characters" is a process in which different characters are used depending on the user's selection, and the characters provide dialogue with the user.

[1541] "Providing product information in a virtual store" is the process by which a character in a virtual environment provides information about products to users.

[1542] "Improving the user's interaction experience" is a process aimed at qualitatively improving the experience that users have through natural interactions with characters.

[1543] This program is a system that allows users to obtain product information through natural dialogue with characters in a virtual store and assists the purchasing process. The system is composed of a server, a terminal, and user operations.

[1544] server

[1545] The server has the following responsibilities:

[1546] 1. Collecting and Learning Character Information:

[1547] The server collects dialogue and voice data provided by character rights holders. The collected data is preprocessed to remove noise and check grammar. Based on this preprocessed data, the generative AI model learns the character's unique dialogue and personality. The learning results are stored in a database.

[1548] 2. Generating and sending conversation data:

[1549] The server receives voice input from the device and converts it into text using a speech recognition system. The converted text is then analyzed using natural language processing (NLP) to understand the user's intent and generate an appropriate response, which is then sent back to the device.

[1550] Terminal

[1551] The terminal has the following responsibilities:

[1552] 1. Capture voice input and convert it to text:

[1553] It takes the user's voice input and captures it through a microphone, which sends the voice data to a speech recognition system and converts it into text data.

[1554] 2. Sending a conversation request and receiving a response:

[1555] The converted text data is sent to the server, and a response is waited for from the server. The terminal again receives the response data sent back from the server.

[1556] 3. Transcribing and playing back response data:

[1557] The received text data is converted into voice data using a speech synthesis system, and the generated voice data is played back through the device's speaker, allowing the user to listen to it.

[1558] user

[1559] The user operates the following functions:

[1560] 1. Character selection and conversation start:

[1561] Users use a terminal application to select their favorite character, which is stored on the terminal, and can then start a conversation by speaking to the selected character.

[1562] 2. Receiving the response:

[1563] The user receives a voice response from the character, and if they wish to continue the conversation, they repeat the same process.

[1564] Hardware and software used

[1565] Hardware: smartphone, microphone, speaker

[1566] Software: Python, SpeechRecognition library, Transformers library, gTTS (Google Text-to-Speech)

[1567] Specific examples

[1568] When a user asks the figure, "What products do you recommend?", the following process takes place:

[1569] The device captures the voice input, converts it into text data, and sends it to the server. The server analyzes the text data, generates an appropriate response, such as "The recommended product is the latest smartphone," and sends it back to the device as text data. The device then converts the received text data into voice data, and the figure replies, "The recommended product is the latest smartphone." In this way, the user enjoys a realistic dialogue with the character.

[1570] Prompt Sentence Examples

[1571] 1. "What products do you recommend?"

[1572] 2. "Please tell me the features of this product."

[1573] 3. "Do you have it in stock?"

[1574] 4. "When is the next sale?"

[1575] This allows the system to enable users to obtain product information through natural interactions with characters in the virtual store, improving the user experience.

[1576] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1577] Step 1:

[1578] The user speaks into the device's microphone, and the user's voice input is captured through the device's microphone.

[1579] Input: User's voice

[1580] Output: Audio data

[1581] Step 2:

[1582] The device sends the captured voice data to a speech recognition system that converts it into text using the SpeechRecognition library.

[1583] Input: Audio data

[1584] Output: Text data

[1585] Step 3:

[1586] The text data is transmitted from the terminal to the server.

[1587] Input: Text data

[1588] Output: Text data sent to the server

[1589] Step 4:

[1590] The server analyzes the received text data using natural language processing (NLP). Specifically, a generative AI model using the Transformers library analyzes the text data and understands the user's intent.

[1591] Input: Text data sent to the server

[1592] Output: Analysis results

[1593] Step 5:

[1594] The server generates an appropriate response based on the analysis results, using a generative AI model to generate the best response to the user's question or request.

[1595] Input: Analysis results

[1596] Output: Response text data

[1597] Step 6:

[1598] The generated response text data is transmitted from the server to the terminal.

[1599] Input: Response text data

[1600] Output: Response text data sent to the terminal

[1601] Step 7:

[1602] The device sends the received response text data to a speech synthesis system, which converts it into voice data. The speech synthesis system uses gTTS (Google Text-to-Speech).

[1603] Input: Response text data sent to the terminal

[1604] Output: Response audio data

[1605] Step 8:

[1606] The generated audio data is played back through the device's speaker, allowing the user to hear it.

[1607] Input: Response audio data

[1608] Output: The response the user hears

[1609] To summarize the specific operations of each step, the user speaks a question into the device, the speech is converted into text and sent to the server, the server analyzes the text data, generates an appropriate response and sends it back to the device, and the device converts the received text response into speech and plays it back to the user, achieving natural dialogue in real time.

[1610] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1611] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with characters in anime or games in real time. The program processing of this system will be specifically described below.

[1612] server

[1613] 1. Collecting and Learning Character Information:

[1614] The server collects dialogue and voice data provided by character rights holders. This data is pre-processed to remove unwanted noise. Based on the pre-processed data, the generative AI model learns the character's unique dialogue and personality.

[1615] 2. Receiving and parsing voice input from the user:

[1616] When a user makes a voice input, the data is sent via the terminal to the server, which receives the voice data and converts it into text data using a voice recognition system.

[1617] 3. Sentiment Analysis of Text Data:

[1618] The server sends the converted text data to the emotion engine, which analyzes the user's emotions and determines their emotional state from the tone and content of their voice.

[1619] 4. Generating an appropriate response:

[1620] Based on the results of the emotion analysis, the server uses a generative AI model to generate an appropriate response. For example, if the user looks sad, the character will respond with comforting words.

[1621] 5. Sending response data:

[1622] The generated response data is sent to the terminal and a reply is sent to the user.

[1623] Terminal

[1624] 1. Capture voice input and convert it to text:

[1625] The device uses a microphone to capture the user's voice, which is then sent to a voice recognition system and converted into text.

[1626] 2. Sending the converted data and receiving the response:

[1627] The terminal transmits the converted text data to the server, waits for response data from the server, and receives the text data returned from the server.

[1628] 3. Transcribing and playing back response data:

[1629] The device sends the received text data to a speech synthesis system, converts it into voice data, and plays the generated voice data through the figure's speaker to respond to the user.

[1630] user

[1631] 1. Character selection and conversation start:

[1632] Users use the terminal application to select their favorite character, and when they speak to the selected character, the figure captures their voice.

[1633] 2. Receiving responses and reflecting emotions:

[1634] The figure responds with a voice response that is based on the user's emotional state, and the emotion engine adjusts the response to take the user's state into account, allowing for more natural and meaningful interactions.

[1635] Specific examples

[1636] For example, if a user says, "Today was not a very fun day," the following process occurs: The device captures the voice data, converts it into text data, and sends it to the server. The server analyzes the text data with an emotion engine and determines that the user is sad. The generative AI model generates comforting words such as "What happened? Thanks for telling me," and sends the text data back to the device. The device converts this data into speech, and the figure replies, "What happened? Thanks for telling me." In this way, the user can continue the dialogue with the character in a natural way.

[1637] Through these processing steps, the system achieves realistic and natural dialogue between the user and the character. The incorporation of an emotion engine generates appropriate responses based on the user's emotions, further improving the quality of the user experience.

[1638] The processing flow will be explained below.

[1639] server

[1640] Step 1:

[1641] The server collects dialogue and voice data provided by character rights holders, including detailed dialogue for specific characters and voice actor recordings.

[1642] Step 2:

[1643] The collected audio data is preprocessed to remove noise and equalize the volume. Noise removal is necessary to remove unwanted background sounds from the audio data.

[1644] Step 3:

[1645] Analyze the pre-processed data and convert the speech into text data. Use a speech recognition system to efficiently convert the collected speech information into text.

[1646] Step 4:

[1647] The text data is used to further train the generative AI model, which then incorporates the character's unique lines and personality traits and updates its parameters.

[1648] Step 5:

[1649] The learning results are stored in a database to enable quick responses to user requests.

[1650] Step 6:

[1651] When a user sends a voice input to the server via the terminal, the server receives it and converts the voice data into text data using a voice recognition system.

[1652] Step 7:

[1653] The converted text data is analyzed using a natural language processing (NLP) engine to understand the user's intent.

[1654] Step 8:

[1655] Analyze user sentiment from text data using an emotion engine, which identifies emotions from the user's phrasing and tone of voice.

[1656] Step 9:

[1657] Based on the analysis results, an appropriate response is generated using a generative AI model, which is then structured as text data.

[1658] Step 10:

[1659] The generated text data is sent to the terminal and a reply is given to the user, which is tailored to the user's emotions.

[1660] Terminal

[1661] Step 1:

[1662] When a user speaks to the figure, the device captures the voice with a microphone, and the captured voice data is processed in real time.

[1663] Step 2:

[1664] The captured voice data is sent to a voice recognition system and converted into text data using voice recognition software.

[1665] Step 3:

[1666] The converted text data is sent to the server and a response is awaited from the server, a necessary step to prepare the server for processing the data.

[1667] Step 4:

[1668] Receives the response data received from the server in text format. This data was generated by the server based on the user's request.

[1669] Step 5:

[1670] The received text data is sent to a speech synthesis system and converted into voice data. Through the speech synthesis process, the text data is turned into voice.

[1671] Step 6:

[1672] The generated voice data is played through the figure's speaker, and the figure uses this voice data to respond to the user.

[1673] user

[1674] Step 1:

[1675] Users use a terminal application to select their favorite character, and the selection information is saved on the terminal.

[1676] Step 2:

[1677] Speak to your chosen character and the figure will capture your voice, which is processed in real time.

[1678] Step 3:

[1679] Wait for a response from the figure. The user simply waits while data is processed between the server and the device.

[1680] Step 4:

[1681] The user can continue the conversation by receiving voice responses from the figure. The responses correspond to the user's emotions, allowing for natural dialogue.

[1682] Examples:

[1683] For example, if a user says to a figurine, "You seem tired and down lately," the following process takes place: The device captures the voice data and converts it into text data using a voice recognition system. This text data is sent to the server, where an emotion engine analyzes the emotion "tired." Based on the analysis results, the generative AI model generates a response such as "Are you OK? You should get some rest," and sends it back to the device as text data. The device converts this data into voice, and the figurine responds, "Are you OK? You should get some rest." In this way, the user can continue to have a natural conversation with the character.

[1684] Through these specific processing steps, the system achieves realistic and natural dialogue between users and characters. The built-in emotion engine generates appropriate responses based on the user's emotions, further improving the user experience and providing new business opportunities for character rights holders.

[1685] Example 2

[1686] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1687] Conventional communication systems are unable to generate appropriate responses based on the user's emotional state, resulting in conversations that tend to be one-sided and mechanical. Furthermore, they lack the means to learn character-specific lines and personalities, resulting in a poor user experience. There is a need for a communication system that can solve these issues and realize more natural and emotionally sensitive conversations.

[1688] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1689] In this invention, the server includes means for acquiring voice input from a user, means for converting the voice input into text data, means for analyzing the text data with an emotion analysis engine and identifying the user's emotional state, means for generating an appropriate response using a generative AI model based on the results of the emotion analysis, means for converting the generated response from text data into voice data, and means for outputting the voice data to the user, thereby enabling natural and meaningful dialogue that is in tune with the user's emotional state.

[1690] "Means for obtaining voice input from a user" refers to a device or method for collecting voice data uttered by a user using a microphone or voice input device.

[1691] The "means for converting speech input into text data" refers to a device or method that uses a speech recognition system to convert acquired speech data into text information.

[1692] "Means for analyzing text data using an emotion analysis engine and identifying a user's emotional state" refers to a device or method that uses text data as input and analyzes and identifies a user's emotional state (e.g., joy, sadness, anger, etc.).

[1693] "Means for generating an appropriate response using a generative AI model based on the results of emotion analysis" refers to a device or method that uses a generative AI model to create a natural and appropriate response, taking into account the user's emotional state obtained through emotion analysis.

[1694] The "means for converting the generated response from text data to voice data" refers to a device or method for converting a response expressed in text data into voice data using a voice synthesis system.

[1695] "Means for outputting audio data to a user" means a device or method for communicating the generated audio data to a user through a speaker or other audio output device.

[1696] "Means for learning character-specific lines and personality information" refers to the process of using a machine learning model to learn given lines and personality information to reflect the character's personality and speech.

[1697] A "means for providing dialogue with different characters in response to a user's selection" is a device or method that generates responses and dialogue specific to the character selected by the user.

[1698] This invention relates to a communication system that combines an emotion recognition engine with a system that allows users to interact with anime and game characters in real time. This system is realized by the roles of a server, terminals, and users, and their cooperation.

[1699] Server Features

[1700] The server collects character information, analyzes voice input, analyzes emotions, generates responses, and transmits response data.

[1701] 1. Collecting and Learning Character Information:

[1702] The server collects dialogue and voice data provided by character rights holders. This data is preprocessed using Python and LibROSA to remove noise. The preprocessed data is then trained by a generative AI model (e.g., GPT-3). This training process allows the system to analyze the character's unique dialogue and personality, resulting in natural-sounding dialogue.

[1703] 2. Receive and analyze user voice input:

[1704] The user's speech is captured by the device and sent to a server, which converts the speech data into text using a speech recognition system, such as Google Speech-to-Text.

[1705] 3. Emotion analysis:

[1706] The server then sends the converted text data to IBM Watson's emotion analysis engine to analyze the user's emotional state. Through this process, the user's emotions (such as joy, sadness, anger, etc.) are identified.

[1707] 4. Generate response:

[1708] Based on the results of the emotion analysis, the generative AI model generates an appropriate response. For example, if the analysis finds that the user is sad, the generative AI model generates comforting words such as "What happened? Thank you for telling me." An example of a prompt sentence could be, "If the user is sad, please generate comforting words."

[1709] 5. Sending response data:

[1710] The server sends the generated response text data to the terminal, which converts this data into voice data and outputs it to the user.

[1711] Device Features

[1712] The terminal is responsible for acquiring user voice, converting it into text data, communicating with the server, and outputting response data as voice.

[1713] 1. Capture voice input and convert it to text:

[1714] The device uses a microphone to capture the user's voice, which is then converted into text data using, for example, Google Speech-to-Text.

[1715] 2. Communication with the server:

[1716] The converted text data is sent to the server, and at the same time, a process is performed to wait for and receive response data from the server.

[1717] 3. Transcribing and playing back response data:

[1718] The device receives the text data returned from the server and converts it into voice data using a speech synthesis system such as Amazon Polly, which then responds to the user through the speaker.

[1719] User Actions

[1720] Users can select a character using a terminal application and begin interacting with that character. The figure's responses are adjusted according to the user's emotional state, allowing for more natural and rich interactions.

[1721] 1. Character selection and dialogue initiation:

[1722] Users select their favorite character through the terminal application, and then speak to the character to activate the dialogue system.

[1723] 2. Receiving responses and reflecting emotions:

[1724] You can receive voice responses from the figure and enjoy natural dialogue that reflects your own emotional state.

[1725] Specific examples

[1726] For example, if a user says, "Today was not a very fun day," the device captures the voice data and converts it into text using Google Speech-to-Text. The server analyzes the text data using IBM Watson's emotion analysis engine and determines that the user is sad. The generative AI model generates a response, "What happened? Thank you for talking," based on the prompt, "If the user is sad, please generate words of comfort." The device converts this response into speech using Amazon Polly, and the figure replies, "What happened? Thank you for talking."

[1727] Through this process, the system achieves realistic and natural dialogue between the user and the character. Appropriate responses generated based on the analysis results of the emotion engine further improve the quality of the dialogue.

[1728] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1729] Step 1: Gather and learn about the character

[1730] The server collects dialogue and voice data provided by character rights holders. The voice data is preprocessed and noise-removed using Python and LibROSA. The input is the provided character's dialogue and voice data, and the output is clean voice data with noise removed.

[1731] Specific operation: The server loads the collected audio files with LibROSA, performs noise filtering using FFT (Fast Fourier Transform), and generates a preprocessed dataset.

[1732] Step 2: Training the generative AI model

[1733] The server trains a generative AI model (e.g., GPT-3) based on the preprocessed data. The input is the preprocessed dataset, and the output is a model that has learned the character's unique lines and personality.

[1734] How it works: The server uses the dataset to train the AI ​​model, building a model that can generate responses that reflect the character's lines and personality traits.

[1735] Step 3: Getting voice input

[1736] The device captures the user's voice as they speak into the microphone. The input is the user's voice data, and the output is the captured raw audio file.

[1737] What it does: The device's microphone captures the user's speech and records it in real time.

[1738] Step 4: Speech to text

[1739] The device sends the captured voice data to a speech recognition system such as Google Speech-to-Text and converts it into text data. The input is the user's voice data, and the output is the corresponding text data.

[1740] Specific operation: The device sends the audio file to the cloud API and receives the returned text data.

[1741] Step 5: Sentiment analysis of text data

[1742] The server receives the text data sent from the device and performs emotion analysis using IBM Watson's emotion analysis engine. The input is the text data, and the output is the analyzed emotional state.

[1743] Specific operation: The server runs the text data through a sentiment analysis engine, resulting in the user's emotion (e.g., joy, sadness, anger).

[1744] Step 6: Generate a response

[1745] The server generates an appropriate response using a generative AI model based on the results of the sentiment analysis. The input is the sentiment analysis result and a prompt, and the output is a generated text response. An example of a prompt is "If the user is sad, please generate comforting words."

[1746] Specific operation: The server inputs a prompt to the generative AI model: "If the user is sad, please generate words of comfort." The model generates a response text such as "What happened? Thank you for telling me."

[1747] Step 7: Transcribing response data

[1748] The device sends the text data sent from the server to a speech synthesis system such as Amazon Polly and converts it into voice data. The input is the response text data, and the output is the synthesized voice data.

[1749] Specific operation: The device sends text data to the speech synthesis API and receives the generated audio file.

[1750] Step 8: Outputting audio data

[1751] The device outputs voice data to the user through a speaker. The input is synthesized voice data, and the output is the voice played from the speaker.

[1752] What happens: The device speaker plays the audio file and the user listens to it.

[1753] Through these processing steps, the system enables natural and emotionally responsive interactions between the user and the character.

[1754] (Application example 2)

[1755] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1756] In conventional communication systems, the dialogue between the user and the character is mechanical and one-way, rather than emotionally responsive. Therefore, there is a need to realize more natural and meaningful dialogue. Furthermore, if appropriate product suggestions can be made based on the user's emotional state, the customer experience can be improved.

[1757] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring voice input from the user, means for converting the voice input into text data, means for recognizing emotions based on the analyzed text data, and means for generating a response and making product suggestions based on the user's emotional state. This makes the dialogue between the user and the character natural and meaningful according to emotions, and further enables appropriate product suggestions based on the user's emotional state.

[1758] A "user" is a person who uses this system to interact with characters.

[1759] "Voice input" refers to the voice data that a user speaks to the system through a microphone.

[1760] "Text data" refers to character data converted by a voice recognition system based on voice input.

[1761] "Analysis" refers to the process of analyzing text data using an emotion recognition engine or similar to understand the user's emotions and intentions.

[1762] "Emotion recognition" is the process of identifying a user's emotional state through analysis.

[1763] "Generating an appropriate response" is the process of generating a reply to establish a dialogue based on the user's emotional state.

[1764] "Product suggestion" is the act of recommending appropriate products based on the user's emotional state and the content of their conversation.

[1765] "Voice data" refers to data that has been converted back into voice format from text data using a voice synthesis system.

[1766] "Output" refers to the act of transmitting the generated audio data to the user through a speaker.

[1767] A "system" is a collection of hardware and software necessary to realize interaction between a user and a character.

[1768] The system for implementing this invention utilizes emotion analysis data to provide appropriate responses and product recommendations in a virtual store where users can interact with anime and game characters. The detailed configuration of this system and the method for realizing it are described below.

[1769] System Configuration

[1770] 1. Server

[1771] Character information gathering and learning:

[1772] The server collects and preprocesses the dialogue and voice data provided by the character rights holders. After preprocessing such as noise removal, the generative AI model learns the character's unique dialogue and personality.

[1773] Receiving and parsing voice input:

[1774] Users input their voice into the virtual store, which is then received by the server, where the voice data is converted into text data using a voice recognition system.

[1775] Sentiment analysis of text data:

[1776] The server sends the analyzed text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) to identify the user's emotion.

[1777] Generate appropriate responses and product suggestions:

[1778] Based on the results of the emotion analysis, a generative AI model (e.g., OpenAI GPT-3) is used to generate appropriate responses, and in addition, product suggestions are made based on the user's emotional state.

[1779] Sending response data:

[1780] The generated response and product proposal are transmitted to the terminal as text data.

[1781] 2. Terminal

[1782] Capture voice input and convert it to text:

[1783] Voice data is captured by the user's microphone and converted into text data using a voice recognition system (e.g., Google Speech-to-Text).

[1784] Sending the transformed data and receiving the response:

[1785] The converted text data is sent to the server, and response data from the server is waited for and received.

[1786] Transcribing and playing response data:

[1787] The received text data is sent to a speech synthesis system (e.g., Amazon Polly) and converted into voice data. The generated voice data is played back through the speaker to respond to the user.

[1788] 3. Users

[1789] Character selection and conversation initiation:

[1790] Users can use the virtual store application, select their favorite character, and start interacting with them. When they speak, the microphone captures the sound and sends it to the device.

[1791] Receiving responses and reflecting emotions:

[1792] The device receives voice feedback responses, which are tailored to the user's emotional state based on an emotion recognition engine.

[1793] Examples and prompts

[1794] For example, if a user says "I'm not feeling well today" in a virtual store, the following happens:

[1795] 1. The device captures the voice data, converts it into text data, and sends it to the server.

[1796] 2. The server performs speech recognition and emotion analysis and determines that the text data "I'm not feeling well" represents a sad emotion.

[1797] 3. The server's generated AI model generates data suggesting products to lift the mood (e.g., pleasant-smelling aromas) along with a response such as, "How's your day? Tell us if you'd like."

[1798] 4. The response data and product suggestions are sent to the terminal and converted into voice by a voice synthesis system.

[1799] 5. The user is prompted with the message, "How's your day going? Let me know if you'd like." along with suggestions for aroma products.

[1800] Prompt Sentence Examples

[1801] "You are a character in a virtual store. A user says, 'I'm feeling down today.' You encourage the user and suggest products that will lift their spirits."

[1802] In this way, this system provides natural and meaningful responses and product suggestions that correspond to the user's emotions through dialogue between the user and the character.

[1803] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1804] Step 1:

[1805] Users simply launch the virtual store application and select their favorite character, which then appears on the screen and can begin interacting with it.

[1806] Step 2:

[1807] The user speaks to the character, saying, "I'm not feeling well today." The device captures this voice data through the microphone and receives it as input.

[1808] Step 3:

[1809] The device uses a voice recognition system (e.g., Google Speech-to-Text) to convert the acquired voice data into text data. At this time, the device processes the voice recognition API and obtains the text data as output.

[1810] Step 4:

[1811] The terminal sends the obtained text data to the server. The text data is transferred as input to the server via the network, and the server receives it.

[1812] Step 5:

[1813] The server sends the received text data to an emotion recognition engine (e.g., IBM Watson, Microsoft Azure Text Analytics) for emotion analysis. The text data is passed as input to the analysis engine, and the analyzed user's emotional state is obtained as output.

[1814] Step 6:

[1815] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-3) based on the results of the emotion analysis. At this time, the prompt sentence is given as input to the generative AI model, and response text data that matches the user's emotion is obtained as output.

[1816] Step 7:

[1817] In addition to the generated responses, the server also generates product suggestions based on the user's emotional state. Using the emotion analysis results as input, the server uses a generative AI model to generate text data containing product suggestions as output.

[1818] Step 8:

[1819] The server transmits text data including the response and the product proposal to the terminal, and transmits the text data generated as input to the terminal as output via the network.

[1820] Step 9:

[1821] The device sends the received text data to a speech synthesis system (e.g., Amazon Polly) and converts it into voice data. Text data is passed as input to the speech synthesis API, and the generated voice data is obtained as output.

[1822] Step 10:

[1823] The generated voice data is played from the device's speaker to deliver a response to the user. Specifically, the device outputs the generated voice data to the speaker, and the user listens to it.

[1824] Step 11:

[1825] The user receives responses and suggestions from the character, and decides whether to continue the conversation or consider a product, thus creating a natural and emotionally relevant continuous interaction between the user and the character.

[1826] Through the above processing steps, the user can have a rich interaction experience with the character and can also receive product suggestions according to their emotional state.

[1827] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1828] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1829] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1830] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1831] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1832] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1833] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1834] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1835] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1836] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1837] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1838] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1839] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1840] 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.

[1841] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1842] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1843] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1844] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1845] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1846] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1847] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1848] The following is further disclosed regarding the above embodiment.

[1849] (Claim 1)

[1850] [a means for obtaining voice input from a user; and

[1851] [Means for converting voice input into text data;

[1852] [means for analyzing the text data and generating an appropriate response;

[1853] [Means for converting the generated response from text data to audio data; and

[1854] [means for outputting audio data to a user;

[1855] A communication system including:

[1856] (Claim 2)

[1857] [The system of claim 1 that learns character-specific lines and voice information.

[1858] (Claim 3)

[1859] The system of claim 1, wherein the system provides interaction with different characters depending on the user's selection.

[1860] "Example 1"

[1861] (Claim 1)

[1862] [a means for obtaining voice input from a user; and

[1863] [Means for converting voice input into text data;

[1864] [means for analyzing the text data and generating an appropriate response;

[1865] [Methods for learning character-specific lines and personalities using generative AI models,

[1866] [Means for converting the generated response from text data to audio data; and

[1867] [means for outputting audio data to a user;

[1868] A system including:

[1869] (Claim 2)

[1870] [The system of claim 1, which learns character-specific dialogue and voice information to train a generative AI model.

[1871] (Claim 3)

[1872] The system of claim 1, wherein the system provides interaction with different characters depending on the user's selection.

[1873] "Application Example 1"

[1874] (Claim 1)

[1875] [a means for obtaining voice input from a user; and

[1876] [Means for converting voice input into text data;

[1877] [means for analyzing the text data and generating an appropriate response;

[1878] [Means for converting the generated response from text data to audio data; and

[1879] [means for outputting audio data to a user;

[1880] [Means for characters to provide product information in virtual stores and improve user interaction experiences;

[1881] A system including:

[1882] (Claim 2)

[1883] [The system of claim 1 that learns character-specific lines and voice information.

[1884] (Claim 3)

[1885] The system of claim 1, wherein the system provides interaction with different characters depending on the user's selection.

[1886] "Example 2: Combining Emotion Engines"

[1887] (Claim 1)

[1888] [a means for obtaining voice input from a user; and

[1889] [Means for converting voice input into text data;

[1890] [Means of analyzing text data with an emotion analysis engine to identify the user's emotional state;

[1891] [Means for generating appropriate responses using a generative AI model based on the results of sentiment analysis; and

[1892] [Means for converting the generated response from text data to audio data; and

[1893] [means for outputting audio data to a user;

[1894] A system including:

[1895] (Claim 2)

[1896] [The system of claim 1 that learns character-specific lines and personality information.

[1897] (Claim 3)

[1898] The system of claim 1, wherein the system provides interaction with different characters depending on the user's selection.

[1899] "Application example 2 when combining emotion engines"

[1900] (Claim 1)

[1901] [a means for obtaining voice input from a user; and

[1902] [Means for converting voice input into text data;

[1903] [means for analyzing the text data and generating an appropriate response;

[1904] [Means for recognizing emotions based on analyzed text data;

[1905] [Means for generating responses and making product suggestions based on the emotional state of the user;

[1906] [Means for converting the generated response from text data to audio data; and

[1907] [means for outputting audio data to a user;

[1908] A system including:

[1909] (Claim 2)

[1910] [The system of claim 1 that learns character-specific lines and voice information.

[1911] (Claim 3)

[1912] The system of claim 1, wherein the system provides interaction with different characters depending on the user's selection. [Explanation of symbols]

[1913] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for obtaining voice input from a user; means for converting voice input into text data; means for analyzing the text data and generating an appropriate response; means for converting the generated response from text data to audio data; means for outputting audio data to a user; A communication system including:

2. 10. The system of claim 1, wherein the system learns character-specific dialogue and voice information.

3. The system of claim 1 , wherein the system provides interaction with different characters depending on the user's selection.

Images