System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a user's input in voice or text form; means for cooperating with a server comprising a generative AI model for analyzing the user's input; means for applying an emotion recognition algorithm to recognize the user's emotion; means for generating an optimal suggestion or response based on the user's input and emotion; and means for audibly or visually providing the generated suggestion or response to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional humanoid robots have limited capabilities for dialogue and interaction with users, making it difficult to achieve natural communication. Furthermore, their low accuracy in emotion recognition makes it difficult for them to provide appropriate suggestions and responses to users. Furthermore, their lack of multilingual support limits their international use. There is a need to solve these issues and provide a more natural and personalized user experience. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for accepting user input in the form of voice or text, a means for linking with a server that includes a generative AI model for analyzing the user input, a means for recognizing the user's emotions by applying an emotion recognition algorithm, a means for generating optimal suggestions or responses based on the user's input and emotion data, and a means for providing the generated suggestions or responses to the user via voice or visually. This enables natural dialogue and appropriate, personalized suggestions and responses to the user. Furthermore, the generative AI model's ability to understand and analyze text, voice, images, and video enables multilingual support and the provision of diverse information. Furthermore, the user experience can be improved by accurately grasping the user's emotions through an emotion recognition algorithm and responding accordingly.

[0006] "User Input" means information provided by a User in the form of voice or text.

[0007] "Audio or text format" refers to the method by which a user provides information in the form of audio or text data.

[0008] "Generative AI model" refers to an artificial intelligence model that uses machine learning algorithms to understand and analyze text, audio, images, and video to generate appropriate output.

[0009] A "server" is a computer system that runs a generative AI model, processing user input data and generating responses and suggestions.

[0010] "Emotion recognition algorithm" refers to a computational method for identifying emotions by analyzing a user's facial expressions and vocal tone.

[0011] "Emotional Data" refers to information about a user's emotions generated by emotion recognition algorithms.

[0012] "Suggestions and responses" refers to information or instructions provided to the user based on the user's input and analysis results.

[0013] "Audio or visual" refers to formats that provide information using audio output, displays, etc.

[0014] "Natural dialogue" refers to smooth, natural communication between humans and machines.

[0015] "Personalized" refers to responses and suggestions that are optimized specifically for that user.

[0016] "Multilingual" refers to the ability to support multiple languages ​​and communicate effectively with multilingual users.

[0017] "Visual means" refers to methods of visually presenting information to users using displays, graphics, etc. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion-recognition algorithm. Specifically, this system is implemented as follows:

[0040] System configuration

[0041] 1. Accepting user input

[0042] The user asks a question or makes a request via voice or text.

[0043] The terminal (robot) accepts voice input and converts it into text data. It also accepts text input directly.

[0044] 2. Data submission and analysis

[0045] The terminal (robot) sends text data to the server.

[0046] The server inputs the received text data into a generative AI model (a model using a machine learning algorithm) and analyzes the question content.

[0047] Generative AI models understand text, speech, images, and video to generate appropriate responses and suggestions.

[0048] 3. Emotion recognition

[0049] The device (robot) captures the user's facial expressions and tone of voice through a camera and microphone and applies emotion recognition algorithms.

[0050] The terminal generates emotion data and transmits it to the server.

[0051] 4. Generating the Response

[0052] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data.

[0053] 5. User Feedback

[0054] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example by providing an explanation audibly while showing an illustration on the display.

[0055] Specific examples

[0056] 1. Coordination suggestions at apparel shops

[0057] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0058] The terminal (robot) converts this voice into text data and sends it to the server.

[0059] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[0060] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[0061] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it asks a question aloud, "What do you think of this outfit?", while displaying an image of a specific outfit on the screen.

[0062] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0063] 2. Menu suggestions at restaurants

[0064] The user asks verbally, "What are your recommendations for today?"

[0065] The terminal (robot) converts this voice into text data and sends it to the server.

[0066] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[0067] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[0068] The server will then use the emotion data to refine suggestions for celebrations or special occasions. For example, it might suggest, "Today is a special day, so how about this special menu?" while showing visual images.

[0069] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The user types a question or request in voice or text form, for example, "What should I wear today?"

[0073] Step 2:

[0074] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[0075] Step 3:

[0076] The device sends the converted text data to the server, and also sends the audio file if necessary.

[0077] Step 4:

[0078] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0079] Step 5:

[0080] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0081] Step 6:

[0082] The device (robot) captures the user's facial expressions and voice tone with a camera and microphone and applies emotion recognition algorithms to determine, for example, whether the user is nervous or relaxed.

[0083] Step 7:

[0084] The device generates emotion data and transmits it to the server. For example, it transmits data indicating that the user is nervous.

[0085] Step 8:

[0086] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0087] Step 9:

[0088] The server transmits the generated proposal and response data to the terminal, for example, image data of a specific outfit.

[0089] Step 10:

[0090] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[0091] Step 11:

[0092] The user can review the suggestions and ask further questions or make requests, for example, "Are there any more formal options?"

[0093] Step 12:

[0094] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[0095] Example 1

[0096] 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."

[0097] Conventional dialogue systems simply analyze user input, making it difficult to generate responses that take the user's emotions and context into account. Furthermore, the low accuracy of voice input analysis makes it impossible to achieve a dialogue that satisfies the user. Therefore, there is a demand for a system that can have natural dialogue with users and provide optimal suggestions.

[0098] 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.

[0099] In this invention, the server includes means for accepting user input in voice or text format, means for converting the voice input into text data, means including a generative AI model for analyzing the user input, means for analyzing the user's facial expressions and tone of voice by applying an emotion recognition algorithm to generate emotion data, means for generating optimal suggestions and responses based on the user's input and emotion data, and means for providing the generated suggestions and responses to the user audio or visually. This makes it possible to provide natural dialogue and highly accurate suggestions that take the user's emotions and situation into consideration.

[0100] A "means for accepting user input in the form of voice or text" is an interface through which a user can make voice or text questions or requests to the system.

[0101] "Means for converting voice input into text data" refers to voice recognition technology for converting voice data received from a user into text data.

[0102] A "server containing a generative AI model for analyzing user input" is a server that has a model that utilizes machine learning to analyze a user's text data and generate appropriate responses or suggestions.

[0103] "Means of applying emotion recognition algorithms to analyze a user's facial expressions and tone of voice to generate emotion data" refers to algorithms that use a camera or microphone to capture a user's facial expressions and tone of voice and then analyze emotions based on that.

[0104] "Means for generating optimal suggestions and responses based on user input and emotional data" refers to the process of integrating the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[0105] "Means for providing generated suggestions or responses to the user via voice or visually" refers to means for delivering responses obtained from a generative AI model to the user using voice synthesis or a display.

[0106] This invention relates to a system that analyzes user input and generates optimal suggestions and responses by combining a generative AI model with an emotion recognition algorithm.

[0107] System configuration

[0108] 1. Accepting user input

[0109] The user asks a question or makes a request via voice or text, for example, "What kind of clothes would you like to wear today?"

[0110] The terminal (robot) accepts voice input and converts it into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). It can also accept text input directly.

[0111] 2. Data submission and analysis

[0112] The device sends the converted text data to the server using an API request.

[0113] The server inputs the received text data into a generative AI model (e.g., OpenAI GPT-4) and analyzes the question. At this time, the server generates a prompt and passes the data to the generative AI model. An example of a prompt is: "The user is asking about what to wear to dinner. Please give us the best suggestions."

[0114] 3. Emotion recognition

[0115] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, and uses emotion recognition algorithms (e.g., Hume AI) to process this data in real time.

[0116] The device sends the generated emotion data to the server. Specifically, it transmits information obtained as emotion data, such as "the user is nervous" or "relaxed."

[0117] 4. Generating the Response

[0118] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, for example, suggesting casual clothing if the user is relaxed.

[0119] The generated response data is saved in text format and sent to the terminal.

[0120] 5. User Feedback

[0121] The device converts the text response sent by the server into speech and provides it to the user, using text-to-speech software (e.g., Amazon Polly) to generate specific responses.

[0122] At the same time, images and text are displayed on the device's display to visually display the response. For example, an image of an outfit is displayed on the display along with the voice message, "What do you think of this outfit?"

[0123] Specific examples

[0124] 1. Coordination suggestions at apparel shops

[0125] User: "I'm going to dinner with a friend today. Can you tell me what to wear?"

[0126] Terminal: Converts voice into text data and sends it to the server.

[0127] Server: Analyzes questions using a generative AI model (e.g., GPT-4) and generates optimal outfits based on weather information and the latest fashion trends.

[0128] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[0129] Server: Corrects the suggestions based on emotional data, asks aloud "What do you think of this outfit?" as a concrete example, and displays an image on the screen.

[0130] User: Review the suggestions and make a follow-up request: "Do you have a more casual jacket?"

[0131] 2. Menu suggestions at restaurants

[0132] User: "What's your recommendation for today?"

[0133] Terminal: Converts voice into text data and sends it to the server.

[0134] Server: Analyzes the question using a generative AI model (e.g., GPT-4) and generates recommendations based on menu information and seasonal specials.

[0135] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[0136] Server: Corrects suggestions based on emotional data, suggests via voice, "Today is a special day, how about this special menu?" and shows a visual image.

[0137] User: Review the suggestions and ask further questions or make requests.

[0138] The system can engage in natural dialogue with users and provide appropriate and personalized suggestions and responses. Specific operations include voice recognition, data transmission and analysis, emotion recognition, response generation, speech synthesis, and display.

[0139] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0140] Step 1:

[0141] The user asks a question or makes a request in the form of voice or text. In this case, the input from the user is voice data or text data. For example, a voice input might be "What's the weather like today?" The device receives this voice and converts the voice data into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). The converted text data is output.

[0142] Step 2:

[0143] The device sends the converted text data to the server. The sent text data is securely sent to the server via an API request. The server uses a generative AI model (e.g., OpenAI GPT-4) to analyze the received text data. At this point, the server generates a prompt and passes the prompt and text data to the generative AI model. An example of a prompt is: "The user is asking about the weather. Please provide the latest weather information." The analysis result is output.

[0144] Step 3:

[0145] The device uses a camera and microphone to capture the user's facial expressions and voice tone. This captured data is analyzed using emotion recognition algorithms (e.g., Hume AI) to generate emotional data, such as whether the user is nervous or relaxed. The emotional data is output.

[0146] Step 4:

[0147] The device sends the generated emotional data to the server. This emotional data is then integrated with the text data analyzed earlier. The server generates optimal suggestions and responses based on the analysis results of the generative AI model and the emotional data. For example, it adjusts the content to provide weather information in softer language if the user is relaxed. The optimal suggestions and responses are then output.

[0148] Step 5:

[0149] The server saves the generated suggestions and responses as text data and sends this data to the device. The device converts this text response data into speech using speech synthesis software (e.g., Amazon Polly) and provides it to the user. The device also displays the response on the display. Specifically, it displays an image of a sunny day along with a voice such as "Today's weather is sunny." Feedback to the user is then complete.

[0150] (Application example 1)

[0151] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0152] Conventional user interface systems simply provide information in response to user input, making it difficult to make flexible suggestions based on the user's emotions and situation. In particular, when dealing with customers in physical stores, there is a demand for technology that can make appropriate suggestions and guidance based on emotional information such as the user's facial expression and tone of voice, but no system that can achieve this exists.

[0153] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0154] In this invention, the server includes means for accepting user input in voice or text format, means for cooperating with the server including a generative AI model for analyzing the user input, means for recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the user's input and emotion data, means for providing the generated suggestions and responses to the user in voice or visual form, and means for correcting the content of suggestions based on the user's emotions. This makes it possible to understand the user's emotions and provide personalized suggestions and guidance that are more closely aligned with the user's needs, particularly in physical stores.

[0155] "User Input" means any question or request made by a User to the System in the form of speech or text.

[0156] A "generative AI model" is an artificial intelligence model that can understand and analyze data such as text, audio, images, and video using machine learning algorithms.

[0157] An "emotion recognition algorithm" is an algorithm that analyzes a user's facial expressions and tone of voice to generate emotional data for the user.

[0158] "Suggestion and response generation" is the process by which the system creates optimal suggestions and responses based on the user's input and emotional data.

[0159] "Audio or visual presentation means" means a means of conveying generated suggestions or responses to the user audibly or visually displaying them on a device such as a display screen or smart mirror.

[0160] "Emotion-based suggestion adjustment" is the process of taking into account the user's emotional data and adjusting the generated suggestions and responses to suit the user's situation and mood.

[0161] This invention is a system that analyzes user input in brick-and-mortar stores, recognizes emotions, and provides appropriate suggestions and responses. Specifically, it uses a system that combines a generative AI model and an emotion recognition algorithm to generate optimal suggestions and responses based on the user's voice or text input and emotional data, and provides them to the user.

[0162] System configuration

[0163] 1. Accepting user input

[0164] Users can ask questions or make requests via voice or text. Smart mirrors and terminals installed in physical stores accept voice input and convert it into text data, and can also accept text input directly.

[0165] 2. Data submission and analysis

[0166] The smart mirror or device sends text data to a server, which then inputs the received text data into a generative AI model to analyze the question. The generative AI model understands text, audio, images, and video and generates appropriate responses and suggestions.

[0167] 3. Emotion recognition

[0168] The smart mirror or device captures the user's facial expressions and voice tone through a camera and microphone, then applies emotion recognition algorithms to the device, generating emotion data that is then sent to a server.

[0169] 4. Generating the Response

[0170] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, and further adjusts the suggestions based on the user's emotions.

[0171] 5. User Feedback

[0172] The smart mirror or device will generate suggestions and responses and provide them to the user both audibly and visually, for example by providing audio instructions accompanied by diagrams on the display.

[0173] Hardware and software used

[0174] Hardware:

[0175] Microphone: Captures the user's voice input.

[0176] Camera: Captures the user's facial expressions.

[0177] Smart mirror: Capable of displaying images and outputting audio.

[0178] software:

[0179] OpenAI API: Used to understand user questions and generate appropriate suggestions and responses through generative AI models.

[0180] SpeechRecognition: Used to convert voice input into text data.

[0181] EmotionRecognition: The algorithm used to recognize emotions from the user's face.

[0182] Pillow: A library for image processing.

[0183] Specific examples

[0184] For example, consider a case where a user speaks to a smart mirror installed in an apparel shop and asks, "I have dinner plans today, so what clothes should I wear?" The smart mirror converts this speech into text and inputs it into a generative AI model. The generative AI model generates appropriate fashion advice based on the input question. At the same time, the smart mirror captures the user's facial expressions with a camera and generates emotion data using an emotion recognition algorithm.

[0185] The server then adjusts the recommendations to best suit the user's situation based on the fashion advice and emotional data generated by the generative AI model. For example, if the server recognizes that the user is relaxed, it generates a response such as, "How about this chic dress? We also have a more casual option for a more relaxed look."

[0186] The generated suggestions are provided to the user via voice and display. For example, the following prompts are input to the generative AI model:

[0187] Today's user asked: 'I have a dinner appointment tonight, what suitable clothes do you recommend?' Provide a fashion advice considering user's preferences.

[0188] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0189] Step 1:

[0190] Accepting user input

[0191] The user asks a question or makes a request by voice or text. The device (such as a smart mirror) accepts the voice input and converts it into text data. It can also accept text input directly.

[0192] Input: User voice or text input

[0193] Output: Text data

[0194] Specific operation: Captures audio with a microphone and converts it to text using the SpeechRecognition library. Text input is accepted as text data.

[0195] Step 2:

[0196] Sending data

[0197] The terminal transmits the converted text data to the server.

[0198] Input: Text data

[0199] Output: Text data sent to the server

[0200] Specific operation: Use the device's communication function to send the converted text data to the server.

[0201] Step 3:

[0202] Data analysis

[0203] The server inputs the received text data into a generative AI model to analyze the user's question, which then generates appropriate responses and suggestions based on the analysis results.

[0204] Input: Text data sent to the server

[0205] Output: Analysis results from the generative AI model

[0206] Specific operation: Using the OpenAI API, text data is input into the generative AI model, a prompt sentence is created, and analysis results are obtained.

[0207] Step 4:

[0208] emotion recognition

[0209] The device captures the user's facial expressions and voice tone through a camera and microphone, applies emotion recognition algorithms to generate emotion data, and then transmits the emotion data to a server.

[0210] Input: User facial and voice data

[0211] Output: Emotion data

[0212] Specific operations: Capture the user's facial expressions with the camera, apply the EmotionRecognition algorithm to analyze the emotions and generate emotion data, capture the voice tone with the microphone, and apply the emotion recognition algorithm.

[0213] Step 5:

[0214] Generating a response

[0215] The server generates optimal suggestions and responses based on the analysis results of the generative AI model and emotional data, and also adjusts the suggestions based on the user's emotions.

[0216] Input: Analysis results of generative AI model, emotion data

[0217] Output: Best suggestion or response

[0218] Specific operation: Emotional data is integrated into the analysis results obtained by the generative AI model to generate personalized suggestions and responses that take into account the user's emotional state.

[0219] Step 6:

[0220] User feedback

[0221] The final suggestion or response is provided to the user via audio and visual means by the device (e.g., smart mirror).

[0222] Input: Best suggestion or response

[0223] Output: Audio and visual feedback provided to the user

[0224] Specific operation: Suggestions and responses are displayed on the smart mirror's display and communicated to the user using voice output.

[0225] 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.

[0226] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[0227] System configuration

[0228] 1. Accepting user input

[0229] The user asks a question or requests via voice or text, for example, "What should I wear today?"

[0230] The terminal (robot) accepts voice input and converts it into text data. If the input is text, it is accepted as is.

[0231] 2. Data submission and analysis

[0232] The terminal (robot) sends the converted text data to the server.

[0233] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0234] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0235] 3. Emotion recognition

[0236] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0237] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[0238] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[0239] 4. Generating the Response

[0240] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0241] 5. User Feedback

[0242] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[0243] Specific examples

[0244] 1. Coordination suggestions at apparel shops

[0245] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0246] The terminal (robot) converts this voice into text data and sends it to the server.

[0247] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[0248] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[0249] The emotion engine analyzes this emotion information to generate emotion data, which is then sent to the server.

[0250] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it might ask a question aloud, such as, "How about this relaxed, casual outfit?", while showing specific outfit images on the display.

[0251] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0252] 2. Menu suggestions at restaurants

[0253] The user asks verbally, "What are your recommendations for today?"

[0254] The terminal (robot) converts this voice into text data and sends it to the server.

[0255] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[0256] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[0257] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. For example, it suggests "Today is a special day, so how about this special menu?" while showing a visual image.

[0258] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0259] The processing flow will be explained below.

[0260] Step 1:

[0261] The user types a question or request in voice or text form, for example, "What should I wear today?"

[0262] Step 2:

[0263] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[0264] Step 3:

[0265] The terminal (robot) sends the converted text data to the server. In the case of voice input, the audio file is also sent.

[0266] Step 4:

[0267] The server inputs the received text data into a generative AI model and analyzes the content of the user's question. For example, it identifies the intent, such as "I'm looking for clothing suggestions."

[0268] Step 5:

[0269] The generative AI model understands text, speech, images, and video, and gathers information from relevant databases and external APIs, such as the latest fashion trends and weather data, to generate optimal outfits.

[0270] Step 6:

[0271] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, thereby understanding the user's emotional state.

[0272] Step 7:

[0273] The device sends the captured facial and voice data to an emotion engine, which runs emotion recognition algorithms, for example, to analyze whether the user is nervous or relaxed.

[0274] Step 8:

[0275] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it sends data such as "The user is nervous."

[0276] Step 9:

[0277] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting relaxing, casual clothing.

[0278] Step 10:

[0279] The server transmits the generated proposal and response data to the terminal, for example, data including a specific coordinated image and description.

[0280] Step 11:

[0281] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it may ask a question aloud, such as "How about this casual jacket?", while showing an image of a specific outfit on the display.

[0282] Step 12:

[0283] The user can review the suggestions and ask further questions or make requests. For example, "Are there any more formal options?"

[0284] Step 13:

[0285] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[0286] In this way, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses based on emotion recognition.

[0287] Example 2

[0288] 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."

[0289] Conventional dialogue systems have difficulty fully understanding the user's intentions and emotions and providing appropriate responses and suggestions. Furthermore, advanced processing is required to convert voice input into text and analyze user input in real time, but systems that integrate these processes remain limited. Therefore, there is a need for a system that can accurately analyze user input and provide responses that take emotions into account.

[0290] 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.

[0291] In this invention, the server includes means for accepting user input in the form of voice or text, means for using voice recognition software to convert the accepted voice data into text data, means for transmitting the converted text data to the server, means for analyzing the user input in cooperation with the server including the generative AI model, means for capturing the user's facial expressions and tone of voice using a camera and a microphone, means for analyzing and recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the analysis results of the generative AI model and the emotion data, and means for providing the generated suggestions and responses to the user audio or visually, thereby enabling accurate analysis of the user's input and providing personalized responses and suggestions that take emotions into consideration.

[0292] "Voice or textual input means" refers to an interface for receiving voice or textual input from a user and processing that data.

[0293] "Speech recognition software" refers to a program used to convert voice data into text data. A specific example is automatic speech recognition (ASR) technology.

[0294] A "server" refers to a computer system connected to a network for processing requests from multiple clients.

[0295] A "generative AI model" refers to an artificial intelligence algorithm that analyzes user input data and generates appropriate responses or suggestions based on that data. A specific example is a natural language processing model.

[0296] "Means of capturing using a camera and microphone" refers to a method of using a camera and microphone to capture a user's facial expressions and tone of voice.

[0297] "Emotion recognition algorithm" refers to an algorithm that analyzes acquired data (facial expressions and voice tone) to recognize the user's emotional state (e.g., joy, sadness, tension, etc.).

[0298] "Emotion engine" refers to a software or hardware configuration for executing emotion recognition algorithms.

[0299] "Means for generating suggestions or responses" refers to a method for creating appropriate suggestions or responses for a user based on the analysis results and emotional data.

[0300] "Audio or visual means" refers to a method for communicating generated suggestions or responses to the user, such as by audio output or display.

[0301] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. The system receives user input in the form of voice or text, analyzes the input, recognizes the user's emotion, and then provides appropriate responses and suggestions.

[0302] System configuration

[0303] 1. Accepting user input

[0304] Users can enter questions or requests in voice or text format, for example, "What should I wear today?"

[0305] The terminal (robot) receives this voice and converts it into text data using voice recognition software (for example, a voice recognition API). If the input is in text format, it is accepted by the system as is.

[0306] 2. Data submission and analysis

[0307] The terminal transmits the converted text data to the server.

[0308] The server inputs the received text data into a generative AI model (specifically, a natural language processing model) and analyzes the question. For example, it identifies the theme "wearing clothes."

[0309] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data APIs, fashion database APIs). For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0310] 3. Emotion recognition

[0311] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0312] The device sends the acquired data to an emotion engine, which applies emotion recognition algorithms (e.g., machine learning algorithms) to determine, for example, whether the user is tense or relaxed.

[0313] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates information such as "the user is nervous."

[0314] 4. Generating the Response

[0315] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0316] 5. User Feedback

[0317] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example, saying "How about this casual jacket?" and showing an image on the display.

[0318] Specific examples

[0319] Coordination suggestions at apparel shops

[0320] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0321] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[0322] The server inputs the text data into a generative AI model (natural language processing model) to analyze the user's intentions, and then generates optimal outfits based on weather information and the latest fashion trends (fashion database API).

[0323] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional information such as whether the user is nervous or relaxed.

[0324] The emotion engine analyzes the emotion information to generate emotion data, which is then sent to the server.

[0325] Based on the acquired emotional data, the server adjusts its suggestions to match the user's preferences, asking aloud, "How about a relaxed, casual outfit?" while displaying specific outfit images on the screen.

[0326] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0327] Menu suggestions at restaurants

[0328] The user asks verbally, "What are your recommendations for today?"

[0329] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[0330] The server inputs the text data into a generative AI model (natural language processing model) and analyzes the question. It then generates recommended menu items by referencing menu information and seasonal special dishes (special dish database).

[0331] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional data.

[0332] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. It suggests aloud, "Today is a special day, so how about this special menu?" while showing visual images.

[0333] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0334] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0335] Step 1: Accept user input

[0336] The user types in a question or request through voice or text, for example, "What should I wear today?"

[0337] Input: User voice or text data

[0338] The terminal (robot) receives voice using a microphone and converts the voice into text data using voice recognition software (e.g., voice recognition API). If the input is text, it is accepted by the system as is.

[0339] Output: Converted text data

[0340] Specific behavior:

[0341] The device collects the user's voice through a microphone.

[0342] Call the speech recognition API and convert the speech data into text data.

[0343] The converted text data is temporarily stored.

[0344] Step 2: Sending data

[0345] The device then sends the converted text data to the server.

[0346] Input: Text data stored on the device

[0347] Output: Text data sent to the server

[0348] Specific behavior:

[0349] The converted text data is sent to the server via the network.

[0350] Verify that the submission was successful.

[0351] Step 3: Analyze the data

[0352] The server inputs the received text data into a generative AI model (e.g., a natural language processing model) and analyzes the question content.

[0353] Input: Text data

[0354] Output: Parsed question topic (e.g. "Outfit suggestions")

[0355] Specific behavior:

[0356] The server preprocesses the received text data and generates a prompt (e.g., "The user is asking, 'What should I wear today?' Please generate an appropriate response.").

[0357] The prompt sentence is input into the generative AI model, and the model begins analysis.

[0358] As a result of the analysis, the topic of the question and related information are identified.

[0359] Step 4: Gather information

[0360] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data API, fashion database API).

[0361] Input: Question topic, related information retrieved from databases and external APIs

[0362] Output: Analysis results (e.g., optimal outfits based on the latest fashion trends or weather data)

[0363] Specific behavior:

[0364] The generative AI model calls database APIs or external APIs to obtain the necessary information.

[0365] Based on the acquired information, the model performs analysis and generates coordination suggestions.

[0366] Step 5: Emotion Recognition

[0367] The device uses a camera and microphone to capture the user's facial expressions and voice tone.

[0368] Input: Camera video data, audio tone data

[0369] Output: Captured facial expression data and voice tone data

[0370] Specific behavior:

[0371] The device captures the user's facial expressions with a camera and collects voice tones with a microphone.

[0372] The collected facial expression data and voice tone data are temporarily stored.

[0373] Step 6: Analyze the sentiment data

[0374] The device sends the acquired data to the emotion engine and applies an emotion recognition algorithm (e.g., a machine learning algorithm).

[0375] Input: facial expression data, voice tone data

[0376] Output: Emotion data (e.g., "The user is nervous")

[0377] Specific behavior:

[0378] Facial expression data and voice tone data are sent to the emotion engine.

[0379] An emotion recognition algorithm is used to analyze emotions and generate emotion data.

[0380] The generated emotion data is sent to the server.

[0381] Step 7: Generate a response

[0382] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[0383] Input: Analysis results (e.g., outfit suggestions), emotion data

[0384] Output: Optimal suggestion or response (e.g., "Suggestions for relaxing casual clothing")

[0385] Specific behavior:

[0386] Integrate analysis results with emotion data.

[0387] It takes into account the user's emotional state to instruct the generative AI model on how to generate a response.

[0388] Generate optimal proposals and responses and send them to the device.

[0389] Step 8: User feedback

[0390] The terminal (robot) provides the generated suggestions and responses to the user audibly and visually.

[0391] Input: Best suggestion or response

[0392] Output: Audio and visual suggestions and responses provided to the user

[0393] Specific behavior:

[0394] The terminal uses speech synthesis software to audibly output the generated suggestions.

[0395] At the same time, an image of the coordinated outfit is displayed on the display.

[0396] The user can review the information provided and ask further questions or requests, for example, "Do you have a more casual jacket?"

[0397] (Application example 2)

[0398] 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."

[0399] In today's brick-and-mortar stores, it is not easy for users to receive accurate and personalized suggestions when selecting products or using services. Conventional systems have difficulty not only analyzing user input, but also properly recognizing the user's emotional state and making optimal suggestions based on this. This has resulted in the problem of insufficient effective customer service and service provision.

[0400] 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 accepting user input in voice or text format; means including a generative AI model for analyzing the user input; means for recognizing the user's emotion by applying an emotion recognition algorithm; means for generating optimal suggestions and responses based on the user's input and emotion data; means for providing the generated suggestions and responses to the user audio or visually; and means for analyzing the user's facial expressions and tone of voice using a camera and microphone when the user asks a question in a physical store, and correcting the generative AI model based on the emotion data to make optimal suggestions. This enables personalized suggestions and responses based on the user's emotional state.

[0401] "User Input" means a question or request provided by a User in voice or text form.

[0402] A "generative AI model" is an artificial intelligence model that analyzes user input, understands text, voice, images, and video, and generates optimal suggestions and responses.

[0403] The "emotion recognition algorithm" is an algorithm that uses a camera and microphone to analyze a user's facial expressions and tone of voice to understand their emotional state.

[0404] "Emotional data" is data that indicates a user's emotional state as analyzed by an emotion recognition algorithm.

[0405] "Suggestions and responses" refer to optimal information and suggestions derived by the generative AI model based on user input and emotional data.

[0406] "Means for providing audio or visual information to the user" refers to means for conveying generated suggestions or responses to the user through audio output or visual information such as a display.

[0407] A "brick and mortar store" is a physical location where consumers can visit and consume products or services.

[0408] "Camera and microphone" refers to video and audio input devices for capturing the user's facial expressions and voice tone.

[0409] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[0410] System configuration

[0411] 1. Accepting user input

[0412] The device provides an interface through which the user can ask questions or make requests in the form of voice or text, for example, by saying, "What should I wear today?"

[0413] The device accepts voice input and converts it into text data, or accepts text input as is.

[0414] 2. Data submission and analysis

[0415] The terminal transmits the converted text data to the server.

[0416] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0417] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0418] 3. Emotion recognition

[0419] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0420] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[0421] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[0422] 4. Generating the Response

[0423] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0424] 5. User Feedback

[0425] The device will then provide the generated suggestions and responses to the user both audibly and visually, for example, asking a question aloud, "What do you think of this casual jacket?", while showing an image on the display.

[0426] Hardware and software used

[0427] 1. Hardware

[0428] camera

[0429] microphone

[0430] Smart glasses, head-mounted displays, tablets

[0431] 2. Software

[0432] Emotion recognition libraries (e.g., Affectiva)

[0433] Generative AI models (e.g., OpenAI GPT-4)

[0434] Speech processing engine (e.g. Google Speech-to-Text)

[0435] GUI library for display (e.g. Qt)

[0436] Examples of concrete examples and prompts

[0437] Examples:

[0438] When a user asks "What's your recommended outfit for today?" and the emotion recognition system determines that the answer is "Relaxed," the prompt is:

[0439] User Question: What's your go-to outfit for today?

[0440] The user is relaxed.

[0441] Please make appropriate suggestions.

[0442] Thus, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses.

[0443] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0444] Step 1:

[0445] A user inputs a question or request to a terminal in the form of voice or text. This is done through a voice interface or a text input interface. For example, the question "What should I wear today?" is input.

[0446] Input: User voice or text question

[0447] Output: Audio data or text data

[0448] Step 2:

[0449] The device converts the voice input into text data. If the input is voice, a speech processing engine (e.g., Google Speech-to-Text) converts the voice into text. If the input is text, this step is skipped.

[0450] Input: Audio data

[0451] Output: Text data

[0452] Step 3:

[0453] The device sends text data to the server, which receives it and inputs it into the generative AI model.

[0454] Input: Text data

[0455] Output: Request for analysis results

[0456] Step 4:

[0457] The server analyzes the text data and understands the question using a generative AI model (e.g., OpenAI GPT-4). The generative AI model collects and analyzes relevant information to generate optimal suggestions, taking into account, for example, the latest fashion trends and weather data.

[0458] Input: Text data

[0459] Output: Initial proposal data

[0460] Step 5:

[0461] The device uses a camera and microphone to capture the user's facial expressions and voice tone, and this data is sent to an emotion recognition algorithm (e.g., Affectiva) to analyze the user's emotional state.

[0462] Input: User's face image and voice tone

[0463] Output: Emotion data

[0464] Step 6:

[0465] The emotion engine generates emotion data analyzed by an emotion recognition algorithm and sends it to the server, which then provides emotion information such as "the user is nervous."

[0466] Input: Facial image data and voice tone data

[0467] Output: Parsed emotion data

[0468] Step 7:

[0469] The server combines the analysis results of the generated AI model with emotional data to generate optimal suggestions and responses. For example, it may suggest "casual clothing that is comfortable to wear."

[0470] Input: Initial proposal data and emotion data

[0471] Output: Optimized proposal data

[0472] Step 8:

[0473] The device provides the generated suggestions and responses to the user both audibly and visually. Using a voice output engine and a display, the device displays an image on the screen while providing a voice prompt such as, "How about this casual jacket?"

[0474] Input: Optimized proposal data

[0475] Output: Audio audio output and visual display

[0476] In this way, each step processes specific input data and produces an appropriate output based on that data, allowing for personalized suggestions and responses to be provided to the user.

[0477] 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.

[0478] 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.

[0479] 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.

[0480] [Second embodiment]

[0481] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0482] 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.

[0483] 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).

[0484] 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.

[0485] 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.

[0486] 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).

[0487] 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.

[0488] 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.

[0489] 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.

[0490] 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.

[0491] 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.

[0492] 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."

[0493] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion-recognition algorithm. Specifically, this system is implemented as follows:

[0494] System configuration

[0495] 1. Accepting user input

[0496] The user asks a question or makes a request via voice or text.

[0497] The terminal (robot) accepts voice input and converts it into text data. It also accepts text input directly.

[0498] 2. Data submission and analysis

[0499] The terminal (robot) sends text data to the server.

[0500] The server inputs the received text data into a generative AI model (a model using a machine learning algorithm) and analyzes the question content.

[0501] Generative AI models understand text, speech, images, and video to generate appropriate responses and suggestions.

[0502] 3. Emotion recognition

[0503] The device (robot) captures the user's facial expressions and tone of voice through a camera and microphone and applies emotion recognition algorithms.

[0504] The terminal generates emotion data and transmits it to the server.

[0505] 4. Generating the Response

[0506] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data.

[0507] 5. User Feedback

[0508] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example by providing an explanation audibly while showing an illustration on the display.

[0509] Specific examples

[0510] 1. Coordination suggestions at apparel shops

[0511] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0512] The terminal (robot) converts this voice into text data and sends it to the server.

[0513] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[0514] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[0515] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it asks a question aloud, "What do you think of this outfit?", while displaying an image of a specific outfit on the screen.

[0516] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0517] 2. Menu suggestions at restaurants

[0518] The user asks verbally, "What are your recommendations for today?"

[0519] The terminal (robot) converts this voice into text data and sends it to the server.

[0520] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[0521] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[0522] The server will then use the emotion data to refine suggestions for celebrations or special occasions. For example, it might suggest, "Today is a special day, so how about this special menu?" while showing visual images.

[0523] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0524] The processing flow will be explained below.

[0525] Step 1:

[0526] The user types a question or request in voice or text form, for example, "What should I wear today?"

[0527] Step 2:

[0528] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[0529] Step 3:

[0530] The device sends the converted text data to the server, and also sends the audio file if necessary.

[0531] Step 4:

[0532] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0533] Step 5:

[0534] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0535] Step 6:

[0536] The device (robot) captures the user's facial expressions and voice tone with a camera and microphone and applies emotion recognition algorithms to determine, for example, whether the user is nervous or relaxed.

[0537] Step 7:

[0538] The device generates emotion data and transmits it to the server. For example, it transmits data indicating that the user is nervous.

[0539] Step 8:

[0540] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0541] Step 9:

[0542] The server transmits the generated proposal and response data to the terminal, for example, image data of a specific outfit.

[0543] Step 10:

[0544] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[0545] Step 11:

[0546] The user can review the suggestions and ask further questions or make requests, for example, "Are there any more formal options?"

[0547] Step 12:

[0548] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[0549] Example 1

[0550] 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."

[0551] Conventional dialogue systems simply analyze user input, making it difficult to generate responses that take the user's emotions and context into account. Furthermore, the low accuracy of voice input analysis makes it impossible to achieve a dialogue that satisfies the user. Therefore, there is a demand for a system that can have natural dialogue with users and provide optimal suggestions.

[0552] 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.

[0553] In this invention, the server includes means for accepting user input in voice or text format, means for converting the voice input into text data, means including a generative AI model for analyzing the user input, means for analyzing the user's facial expressions and tone of voice by applying an emotion recognition algorithm to generate emotion data, means for generating optimal suggestions and responses based on the user's input and emotion data, and means for providing the generated suggestions and responses to the user audio or visually. This makes it possible to provide natural dialogue and highly accurate suggestions that take the user's emotions and situation into consideration.

[0554] A "means for accepting user input in the form of voice or text" is an interface through which a user can make voice or text questions or requests to the system.

[0555] "Means for converting voice input into text data" refers to voice recognition technology for converting voice data received from a user into text data.

[0556] A "server containing a generative AI model for analyzing user input" is a server that has a model that utilizes machine learning to analyze a user's text data and generate appropriate responses or suggestions.

[0557] "Means of applying emotion recognition algorithms to analyze a user's facial expressions and tone of voice to generate emotion data" refers to algorithms that use a camera or microphone to capture a user's facial expressions and tone of voice and then analyze emotions based on that.

[0558] "Means for generating optimal suggestions and responses based on user input and emotional data" refers to the process of integrating the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[0559] "Means for providing generated suggestions or responses to the user via voice or visually" refers to means for delivering responses obtained from a generative AI model to the user using voice synthesis or a display.

[0560] This invention relates to a system that analyzes user input and generates optimal suggestions and responses by combining a generative AI model with an emotion recognition algorithm.

[0561] System configuration

[0562] 1. Accepting user input

[0563] The user asks a question or makes a request via voice or text, for example, "What kind of clothes would you like to wear today?"

[0564] The terminal (robot) accepts voice input and converts it into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). It can also accept text input directly.

[0565] 2. Data submission and analysis

[0566] The device sends the converted text data to the server using an API request.

[0567] The server inputs the received text data into a generative AI model (e.g., OpenAI GPT-4) and analyzes the question. At this time, the server generates a prompt and passes the data to the generative AI model. An example of a prompt is: "The user is asking about what to wear to dinner. Please give us the best suggestions."

[0568] 3. Emotion recognition

[0569] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, and uses emotion recognition algorithms (e.g., Hume AI) to process this data in real time.

[0570] The device sends the generated emotion data to the server. Specifically, it transmits information obtained as emotion data, such as "the user is nervous" or "relaxed."

[0571] 4. Generating the Response

[0572] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, for example, suggesting casual clothing if the user is relaxed.

[0573] The generated response data is saved in text format and sent to the terminal.

[0574] 5. User Feedback

[0575] The device converts the text response sent by the server into speech and provides it to the user, using text-to-speech software (e.g., Amazon Polly) to generate specific responses.

[0576] At the same time, images and text are displayed on the device's display to visually display the response. For example, an image of an outfit is displayed on the display along with the voice message, "What do you think of this outfit?"

[0577] Specific examples

[0578] 1. Coordination suggestions at apparel shops

[0579] User: "I'm going to dinner with a friend today. Can you tell me what to wear?"

[0580] Terminal: Converts voice into text data and sends it to the server.

[0581] Server: Analyzes questions using a generative AI model (e.g., GPT-4) and generates optimal outfits based on weather information and the latest fashion trends.

[0582] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[0583] Server: Corrects the suggestions based on emotional data, asks aloud "What do you think of this outfit?" as a concrete example, and displays an image on the screen.

[0584] User: Review the suggestions and make a follow-up request: "Do you have a more casual jacket?"

[0585] 2. Menu suggestions at restaurants

[0586] User: "What's your recommendation for today?"

[0587] Terminal: Converts voice into text data and sends it to the server.

[0588] Server: Analyzes the question using a generative AI model (e.g., GPT-4) and generates recommendations based on menu information and seasonal specials.

[0589] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[0590] Server: Corrects suggestions based on emotional data, suggests via voice, "Today is a special day, how about this special menu?" and shows a visual image.

[0591] User: Review the suggestions and ask further questions or make requests.

[0592] The system can engage in natural dialogue with users and provide appropriate and personalized suggestions and responses. Specific operations include voice recognition, data transmission and analysis, emotion recognition, response generation, speech synthesis, and display.

[0593] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0594] Step 1:

[0595] The user asks a question or makes a request in the form of voice or text. In this case, the input from the user is voice data or text data. For example, a voice input might be "What's the weather like today?" The device receives this voice and converts the voice data into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). The converted text data is output.

[0596] Step 2:

[0597] The device sends the converted text data to the server. The sent text data is securely sent to the server via an API request. The server uses a generative AI model (e.g., OpenAI GPT-4) to analyze the received text data. At this point, the server generates a prompt and passes the prompt and text data to the generative AI model. An example of a prompt is: "The user is asking about the weather. Please provide the latest weather information." The analysis result is output.

[0598] Step 3:

[0599] The device uses a camera and microphone to capture the user's facial expressions and voice tone. This captured data is analyzed using emotion recognition algorithms (e.g., Hume AI) to generate emotional data, such as whether the user is nervous or relaxed. The emotional data is output.

[0600] Step 4:

[0601] The device sends the generated emotional data to the server. This emotional data is then integrated with the text data analyzed earlier. The server generates optimal suggestions and responses based on the analysis results of the generative AI model and the emotional data. For example, it adjusts the content to provide weather information in softer language if the user is relaxed. The optimal suggestions and responses are then output.

[0602] Step 5:

[0603] The server saves the generated suggestions and responses as text data and sends this data to the device. The device converts this text response data into speech using speech synthesis software (e.g., Amazon Polly) and provides it to the user. The device also displays the response on the display. Specifically, it displays an image of a sunny day along with a voice such as "Today's weather is sunny." Feedback to the user is then complete.

[0604] (Application example 1)

[0605] 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."

[0606] Conventional user interface systems simply provide information in response to user input, making it difficult to make flexible suggestions based on the user's emotions and situation. In particular, when dealing with customers in physical stores, there is a demand for technology that can make appropriate suggestions and guidance based on emotional information such as the user's facial expression and tone of voice, but no system that can achieve this exists.

[0607] 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.

[0608] In this invention, the server includes means for accepting user input in voice or text format, means for cooperating with the server including a generative AI model for analyzing the user input, means for recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the user's input and emotion data, means for providing the generated suggestions and responses to the user in voice or visual form, and means for correcting the content of suggestions based on the user's emotions. This makes it possible to understand the user's emotions and provide personalized suggestions and guidance that are more closely aligned with the user's needs, particularly in physical stores.

[0609] "User Input" means any question or request made by a User to the System in the form of speech or text.

[0610] A "generative AI model" is an artificial intelligence model that can understand and analyze data such as text, audio, images, and video using machine learning algorithms.

[0611] An "emotion recognition algorithm" is an algorithm that analyzes a user's facial expressions and tone of voice to generate emotional data for the user.

[0612] "Suggestion and response generation" is the process by which the system creates optimal suggestions and responses based on the user's input and emotional data.

[0613] "Audio or visual presentation means" means a means of conveying generated suggestions or responses to the user audibly or visually displaying them on a device such as a display screen or smart mirror.

[0614] "Emotion-based suggestion adjustment" is the process of taking into account the user's emotional data and adjusting the generated suggestions and responses to suit the user's situation and mood.

[0615] This invention is a system that analyzes user input in brick-and-mortar stores, recognizes emotions, and provides appropriate suggestions and responses. Specifically, it uses a system that combines a generative AI model and an emotion recognition algorithm to generate optimal suggestions and responses based on the user's voice or text input and emotional data, and provides them to the user.

[0616] System configuration

[0617] 1. Accepting user input

[0618] Users can ask questions or make requests via voice or text. Smart mirrors and terminals installed in physical stores accept voice input and convert it into text data, and can also accept text input directly.

[0619] 2. Data submission and analysis

[0620] The smart mirror or device sends text data to a server, which then inputs the received text data into a generative AI model to analyze the question. The generative AI model understands text, audio, images, and video and generates appropriate responses and suggestions.

[0621] 3. Emotion recognition

[0622] The smart mirror or device captures the user's facial expressions and voice tone through a camera and microphone, then applies emotion recognition algorithms to the device, generating emotion data that is then sent to a server.

[0623] 4. Generating the Response

[0624] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, and further adjusts the suggestions based on the user's emotions.

[0625] 5. User Feedback

[0626] The smart mirror or device will generate suggestions and responses and provide them to the user both audibly and visually, for example by providing audio instructions accompanied by diagrams on the display.

[0627] Hardware and software used

[0628] Hardware:

[0629] Microphone: Captures the user's voice input.

[0630] Camera: Captures the user's facial expressions.

[0631] Smart mirror: Capable of displaying images and outputting audio.

[0632] software:

[0633] OpenAI API: Used to understand user questions and generate appropriate suggestions and responses through generative AI models.

[0634] SpeechRecognition: Used to convert voice input into text data.

[0635] EmotionRecognition: The algorithm used to recognize emotions from the user's face.

[0636] Pillow: A library for image processing.

[0637] Specific examples

[0638] For example, consider a case where a user speaks to a smart mirror installed in an apparel shop and asks, "I have dinner plans today, so what clothes should I wear?" The smart mirror converts this speech into text and inputs it into a generative AI model. The generative AI model generates appropriate fashion advice based on the input question. At the same time, the smart mirror captures the user's facial expressions with a camera and generates emotion data using an emotion recognition algorithm.

[0639] The server then adjusts the recommendations to best suit the user's situation based on the fashion advice and emotional data generated by the generative AI model. For example, if the server recognizes that the user is relaxed, it generates a response such as, "How about this chic dress? We also have a more casual option for a more relaxed look."

[0640] The generated suggestions are provided to the user via voice and display. For example, the following prompts are input to the generative AI model:

[0641] Today's user asked: 'I have a dinner appointment tonight, what suitable clothes do you recommend?' Provide a fashion advice considering user's preferences.

[0642] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0643] Step 1:

[0644] Accepting user input

[0645] The user asks a question or makes a request by voice or text. The device (such as a smart mirror) accepts the voice input and converts it into text data. It can also accept text input directly.

[0646] Input: User voice or text input

[0647] Output: Text data

[0648] Specific operation: Captures audio with a microphone and converts it to text using the SpeechRecognition library. Text input is accepted as text data.

[0649] Step 2:

[0650] Sending data

[0651] The terminal transmits the converted text data to the server.

[0652] Input: Text data

[0653] Output: Text data sent to the server

[0654] Specific operation: Use the device's communication function to send the converted text data to the server.

[0655] Step 3:

[0656] Data analysis

[0657] The server inputs the received text data into a generative AI model to analyze the user's question, which then generates appropriate responses and suggestions based on the analysis results.

[0658] Input: Text data sent to the server

[0659] Output: Analysis results from the generative AI model

[0660] Specific operation: Using the OpenAI API, text data is input into the generative AI model, a prompt sentence is created, and analysis results are obtained.

[0661] Step 4:

[0662] emotion recognition

[0663] The device captures the user's facial expressions and voice tone through a camera and microphone, applies emotion recognition algorithms to generate emotion data, and then transmits the emotion data to a server.

[0664] Input: User facial and voice data

[0665] Output: Emotion data

[0666] Specific operations: Capture the user's facial expressions with the camera, apply the EmotionRecognition algorithm to analyze the emotions and generate emotion data, capture the voice tone with the microphone, and apply the emotion recognition algorithm.

[0667] Step 5:

[0668] Generating a response

[0669] The server generates optimal suggestions and responses based on the analysis results of the generative AI model and emotional data, and also adjusts the suggestions based on the user's emotions.

[0670] Input: Analysis results of generative AI model, emotion data

[0671] Output: Best suggestion or response

[0672] Specific operation: Emotional data is integrated into the analysis results obtained by the generative AI model to generate personalized suggestions and responses that take into account the user's emotional state.

[0673] Step 6:

[0674] User feedback

[0675] The final suggestion or response is provided to the user via audio and visual means by the device (e.g., smart mirror).

[0676] Input: Best suggestion or response

[0677] Output: Audio and visual feedback provided to the user

[0678] Specific operation: Suggestions and responses are displayed on the smart mirror's display and communicated to the user using voice output.

[0679] 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.

[0680] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[0681] System configuration

[0682] 1. Accepting user input

[0683] The user asks a question or requests via voice or text, for example, "What should I wear today?"

[0684] The terminal (robot) accepts voice input and converts it into text data. If the input is text, it is accepted as is.

[0685] 2. Data submission and analysis

[0686] The terminal (robot) sends the converted text data to the server.

[0687] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0688] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0689] 3. Emotion recognition

[0690] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0691] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[0692] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[0693] 4. Generating the Response

[0694] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0695] 5. User Feedback

[0696] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[0697] Specific examples

[0698] 1. Coordination suggestions at apparel shops

[0699] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0700] The terminal (robot) converts this voice into text data and sends it to the server.

[0701] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[0702] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[0703] The emotion engine analyzes this emotion information to generate emotion data, which is then sent to the server.

[0704] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it might ask a question aloud, such as, "How about this relaxed, casual outfit?", while showing specific outfit images on the display.

[0705] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0706] 2. Menu suggestions at restaurants

[0707] The user asks verbally, "What are your recommendations for today?"

[0708] The terminal (robot) converts this voice into text data and sends it to the server.

[0709] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[0710] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[0711] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. For example, it suggests "Today is a special day, so how about this special menu?" while showing a visual image.

[0712] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0713] The processing flow will be explained below.

[0714] Step 1:

[0715] The user types a question or request in voice or text form, for example, "What should I wear today?"

[0716] Step 2:

[0717] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[0718] Step 3:

[0719] The terminal (robot) sends the converted text data to the server. In the case of voice input, the audio file is also sent.

[0720] Step 4:

[0721] The server inputs the received text data into a generative AI model and analyzes the content of the user's question. For example, it identifies the intent, such as "I'm looking for clothing suggestions."

[0722] Step 5:

[0723] The generative AI model understands text, speech, images, and video, and gathers information from relevant databases and external APIs, such as the latest fashion trends and weather data, to generate optimal outfits.

[0724] Step 6:

[0725] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, thereby understanding the user's emotional state.

[0726] Step 7:

[0727] The device sends the captured facial and voice data to an emotion engine, which runs emotion recognition algorithms, for example, to analyze whether the user is nervous or relaxed.

[0728] Step 8:

[0729] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it sends data such as "The user is nervous."

[0730] Step 9:

[0731] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting relaxing, casual clothing.

[0732] Step 10:

[0733] The server transmits the generated proposal and response data to the terminal, for example, data including a specific coordinated image and description.

[0734] Step 11:

[0735] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it may ask a question aloud, such as "How about this casual jacket?", while showing an image of a specific outfit on the display.

[0736] Step 12:

[0737] The user can review the suggestions and ask further questions or make requests. For example, "Are there any more formal options?"

[0738] Step 13:

[0739] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[0740] In this way, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses based on emotion recognition.

[0741] Example 2

[0742] 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."

[0743] Conventional dialogue systems have difficulty fully understanding the user's intentions and emotions and providing appropriate responses and suggestions. Furthermore, advanced processing is required to convert voice input into text and analyze user input in real time, but systems that integrate these processes remain limited. Therefore, there is a need for a system that can accurately analyze user input and provide responses that take emotions into account.

[0744] 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.

[0745] In this invention, the server includes means for accepting user input in the form of voice or text, means for using voice recognition software to convert the accepted voice data into text data, means for transmitting the converted text data to the server, means for analyzing the user input in cooperation with the server including the generative AI model, means for capturing the user's facial expressions and tone of voice using a camera and a microphone, means for analyzing and recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the analysis results of the generative AI model and the emotion data, and means for providing the generated suggestions and responses to the user audio or visually, thereby enabling accurate analysis of the user's input and providing personalized responses and suggestions that take emotions into consideration.

[0746] "Voice or textual input means" refers to an interface for receiving voice or textual input from a user and processing that data.

[0747] "Speech recognition software" refers to a program used to convert voice data into text data. A specific example is automatic speech recognition (ASR) technology.

[0748] A "server" refers to a computer system connected to a network for processing requests from multiple clients.

[0749] A "generative AI model" refers to an artificial intelligence algorithm that analyzes user input data and generates appropriate responses or suggestions based on that data. A specific example is a natural language processing model.

[0750] "Means of capturing using a camera and microphone" refers to a method of using a camera and microphone to capture a user's facial expressions and tone of voice.

[0751] "Emotion recognition algorithm" refers to an algorithm that analyzes acquired data (facial expressions and voice tone) to recognize the user's emotional state (e.g., joy, sadness, tension, etc.).

[0752] "Emotion engine" refers to a software or hardware configuration for executing emotion recognition algorithms.

[0753] "Means for generating suggestions or responses" refers to a method for creating appropriate suggestions or responses for a user based on the analysis results and emotional data.

[0754] "Audio or visual means" refers to a method for communicating generated suggestions or responses to the user, such as by audio output or display.

[0755] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. The system receives user input in the form of voice or text, analyzes the input, recognizes the user's emotion, and then provides appropriate responses and suggestions.

[0756] System configuration

[0757] 1. Accepting user input

[0758] Users can enter questions or requests in voice or text format, for example, "What should I wear today?"

[0759] The terminal (robot) receives this voice and converts it into text data using voice recognition software (for example, a voice recognition API). If the input is in text format, it is accepted by the system as is.

[0760] 2. Data submission and analysis

[0761] The terminal transmits the converted text data to the server.

[0762] The server inputs the received text data into a generative AI model (specifically, a natural language processing model) and analyzes the question. For example, it identifies the theme "wearing clothes."

[0763] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data APIs, fashion database APIs). For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0764] 3. Emotion recognition

[0765] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0766] The device sends the acquired data to an emotion engine, which applies emotion recognition algorithms (e.g., machine learning algorithms) to determine, for example, whether the user is tense or relaxed.

[0767] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates information such as "the user is nervous."

[0768] 4. Generating the Response

[0769] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0770] 5. User Feedback

[0771] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example, saying "How about this casual jacket?" and showing an image on the display.

[0772] Specific examples

[0773] Coordination suggestions at apparel shops

[0774] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0775] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[0776] The server inputs the text data into a generative AI model (natural language processing model) to analyze the user's intentions, and then generates optimal outfits based on weather information and the latest fashion trends (fashion database API).

[0777] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional information such as whether the user is nervous or relaxed.

[0778] The emotion engine analyzes the emotion information to generate emotion data, which is then sent to the server.

[0779] Based on the acquired emotional data, the server adjusts its suggestions to match the user's preferences, asking aloud, "How about a relaxed, casual outfit?" while displaying specific outfit images on the screen.

[0780] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0781] Menu suggestions at restaurants

[0782] The user asks verbally, "What are your recommendations for today?"

[0783] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[0784] The server inputs the text data into a generative AI model (natural language processing model) and analyzes the question. It then generates recommended menu items by referencing menu information and seasonal special dishes (special dish database).

[0785] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional data.

[0786] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. It suggests aloud, "Today is a special day, so how about this special menu?" while showing visual images.

[0787] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0788] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0789] Step 1: Accept user input

[0790] The user types in a question or request through voice or text, for example, "What should I wear today?"

[0791] Input: User voice or text data

[0792] The terminal (robot) receives voice using a microphone and converts the voice into text data using voice recognition software (e.g., voice recognition API). If the input is text, it is accepted by the system as is.

[0793] Output: Converted text data

[0794] Specific behavior:

[0795] The device collects the user's voice through a microphone.

[0796] Call the speech recognition API and convert the speech data into text data.

[0797] The converted text data is temporarily stored.

[0798] Step 2: Sending data

[0799] The device then sends the converted text data to the server.

[0800] Input: Text data stored on the device

[0801] Output: Text data sent to the server

[0802] Specific behavior:

[0803] The converted text data is sent to the server via the network.

[0804] Verify that the submission was successful.

[0805] Step 3: Analyze the data

[0806] The server inputs the received text data into a generative AI model (e.g., a natural language processing model) and analyzes the question content.

[0807] Input: Text data

[0808] Output: Parsed question topic (e.g. "Outfit suggestions")

[0809] Specific behavior:

[0810] The server preprocesses the received text data and generates a prompt (e.g., "The user is asking, 'What should I wear today?' Please generate an appropriate response.").

[0811] The prompt sentence is input into the generative AI model, and the model begins analysis.

[0812] As a result of the analysis, the topic of the question and related information are identified.

[0813] Step 4: Gather information

[0814] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data API, fashion database API).

[0815] Input: Question topic, related information retrieved from databases and external APIs

[0816] Output: Analysis results (e.g., optimal outfits based on the latest fashion trends or weather data)

[0817] Specific behavior:

[0818] The generative AI model calls database APIs or external APIs to obtain the necessary information.

[0819] Based on the acquired information, the model performs analysis and generates coordination suggestions.

[0820] Step 5: Emotion Recognition

[0821] The device uses a camera and microphone to capture the user's facial expressions and voice tone.

[0822] Input: Camera video data, audio tone data

[0823] Output: Captured facial expression data and voice tone data

[0824] Specific behavior:

[0825] The device captures the user's facial expressions with a camera and collects voice tones with a microphone.

[0826] The collected facial expression data and voice tone data are temporarily stored.

[0827] Step 6: Analyze the sentiment data

[0828] The device sends the acquired data to the emotion engine and applies an emotion recognition algorithm (e.g., a machine learning algorithm).

[0829] Input: facial expression data, voice tone data

[0830] Output: Emotion data (e.g., "The user is nervous")

[0831] Specific behavior:

[0832] Facial expression data and voice tone data are sent to the emotion engine.

[0833] An emotion recognition algorithm is used to analyze emotions and generate emotion data.

[0834] The generated emotion data is sent to the server.

[0835] Step 7: Generate a response

[0836] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[0837] Input: Analysis results (e.g., outfit suggestions), emotion data

[0838] Output: Optimal suggestion or response (e.g., "Suggestions for relaxing casual clothing")

[0839] Specific behavior:

[0840] Integrate analysis results with emotion data.

[0841] It takes into account the user's emotional state to instruct the generative AI model on how to generate a response.

[0842] Generate optimal proposals and responses and send them to the device.

[0843] Step 8: User feedback

[0844] The terminal (robot) provides the generated suggestions and responses to the user audibly and visually.

[0845] Input: Best suggestion or response

[0846] Output: Audio and visual suggestions and responses provided to the user

[0847] Specific behavior:

[0848] The terminal uses speech synthesis software to audibly output the generated suggestions.

[0849] At the same time, an image of the coordinated outfit is displayed on the display.

[0850] The user can review the information provided and ask further questions or requests, for example, "Do you have a more casual jacket?"

[0851] (Application example 2)

[0852] 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."

[0853] In today's brick-and-mortar stores, it is not easy for users to receive accurate and personalized suggestions when selecting products or using services. Conventional systems have difficulty not only analyzing user input, but also properly recognizing the user's emotional state and making optimal suggestions based on this. This has resulted in the problem of insufficient effective customer service and service provision.

[0854] 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 accepting user input in voice or text format; means including a generative AI model for analyzing the user input; means for recognizing the user's emotion by applying an emotion recognition algorithm; means for generating optimal suggestions and responses based on the user's input and emotion data; means for providing the generated suggestions and responses to the user audio or visually; and means for analyzing the user's facial expressions and tone of voice using a camera and microphone when the user asks a question in a physical store, and correcting the generative AI model based on the emotion data to make optimal suggestions. This enables personalized suggestions and responses based on the user's emotional state.

[0855] "User Input" means a question or request provided by a User in voice or text form.

[0856] A "generative AI model" is an artificial intelligence model that analyzes user input, understands text, voice, images, and video, and generates optimal suggestions and responses.

[0857] The "emotion recognition algorithm" is an algorithm that uses a camera and microphone to analyze a user's facial expressions and tone of voice to understand their emotional state.

[0858] "Emotional data" is data that indicates a user's emotional state as analyzed by an emotion recognition algorithm.

[0859] "Suggestions and responses" refer to optimal information and suggestions derived by the generative AI model based on user input and emotional data.

[0860] "Means for providing audio or visual information to the user" refers to means for conveying generated suggestions or responses to the user through audio output or visual information such as a display.

[0861] A "brick and mortar store" is a physical location where consumers can visit and consume products or services.

[0862] "Camera and microphone" refers to video and audio input devices for capturing the user's facial expressions and voice tone.

[0863] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[0864] System configuration

[0865] 1. Accepting user input

[0866] The device provides an interface through which the user can ask questions or make requests in the form of voice or text, for example, by saying, "What should I wear today?"

[0867] The device accepts voice input and converts it into text data, or accepts text input as is.

[0868] 2. Data submission and analysis

[0869] The terminal transmits the converted text data to the server.

[0870] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0871] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0872] 3. Emotion recognition

[0873] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[0874] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[0875] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[0876] 4. Generating the Response

[0877] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0878] 5. User Feedback

[0879] The device will then provide the generated suggestions and responses to the user both audibly and visually, for example, asking a question aloud, "What do you think of this casual jacket?", while showing an image on the display.

[0880] Hardware and software used

[0881] 1. Hardware

[0882] camera

[0883] microphone

[0884] Smart glasses, head-mounted displays, tablets

[0885] 2. Software

[0886] Emotion recognition libraries (e.g., Affectiva)

[0887] Generative AI models (e.g., OpenAI GPT-4)

[0888] Speech processing engine (e.g. Google Speech-to-Text)

[0889] GUI library for display (e.g. Qt)

[0890] Examples of concrete examples and prompts

[0891] Examples:

[0892] When a user asks "What's your recommended outfit for today?" and the emotion recognition system determines that the answer is "Relaxed," the prompt is:

[0893] User Question: What's your go-to outfit for today?

[0894] The user is relaxed.

[0895] Please make appropriate suggestions.

[0896] Thus, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses.

[0897] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0898] Step 1:

[0899] A user inputs a question or request to a terminal in the form of voice or text. This is done through a voice interface or a text input interface. For example, the question "What should I wear today?" is input.

[0900] Input: User voice or text question

[0901] Output: Audio data or text data

[0902] Step 2:

[0903] The device converts the voice input into text data. If the input is voice, a speech processing engine (e.g., Google Speech-to-Text) converts the voice into text. If the input is text, this step is skipped.

[0904] Input: Audio data

[0905] Output: Text data

[0906] Step 3:

[0907] The device sends text data to the server, which receives it and inputs it into the generative AI model.

[0908] Input: Text data

[0909] Output: Request for analysis results

[0910] Step 4:

[0911] The server analyzes the text data and understands the question using a generative AI model (e.g., OpenAI GPT-4). The generative AI model collects and analyzes relevant information to generate optimal suggestions, taking into account, for example, the latest fashion trends and weather data.

[0912] Input: Text data

[0913] Output: Initial proposal data

[0914] Step 5:

[0915] The device uses a camera and microphone to capture the user's facial expressions and voice tone, and this data is sent to an emotion recognition algorithm (e.g., Affectiva) to analyze the user's emotional state.

[0916] Input: User's face image and voice tone

[0917] Output: Emotion data

[0918] Step 6:

[0919] The emotion engine generates emotion data analyzed by an emotion recognition algorithm and sends it to the server, which then provides emotion information such as "the user is nervous."

[0920] Input: Facial image data and voice tone data

[0921] Output: Parsed emotion data

[0922] Step 7:

[0923] The server combines the analysis results of the generated AI model with emotional data to generate optimal suggestions and responses. For example, it may suggest "casual clothing that is comfortable to wear."

[0924] Input: Initial proposal data and emotion data

[0925] Output: Optimized proposal data

[0926] Step 8:

[0927] The device provides the generated suggestions and responses to the user both audibly and visually. Using a voice output engine and a display, the device displays an image on the screen while providing a voice prompt such as, "How about this casual jacket?"

[0928] Input: Optimized proposal data

[0929] Output: Audio audio output and visual display

[0930] In this way, each step processes specific input data and produces an appropriate output based on that data, allowing for personalized suggestions and responses to be provided to the user.

[0931] 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.

[0932] 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.

[0933] 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.

[0934] [Third embodiment]

[0935] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0936] 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.

[0937] 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).

[0938] 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.

[0939] 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.

[0940] 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).

[0941] 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.

[0942] 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.

[0943] 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.

[0944] 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.

[0945] 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.

[0946] 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."

[0947] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion-recognition algorithm. Specifically, this system is implemented as follows:

[0948] System configuration

[0949] 1. Accepting user input

[0950] The user asks a question or makes a request via voice or text.

[0951] The terminal (robot) accepts voice input and converts it into text data. It also accepts text input directly.

[0952] 2. Data submission and analysis

[0953] The terminal (robot) sends text data to the server.

[0954] The server inputs the received text data into a generative AI model (a model using a machine learning algorithm) and analyzes the question content.

[0955] Generative AI models understand text, speech, images, and video to generate appropriate responses and suggestions.

[0956] 3. Emotion recognition

[0957] The device (robot) captures the user's facial expressions and tone of voice through a camera and microphone and applies emotion recognition algorithms.

[0958] The terminal generates emotion data and transmits it to the server.

[0959] 4. Generating the Response

[0960] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data.

[0961] 5. User Feedback

[0962] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example by providing an explanation audibly while showing an illustration on the display.

[0963] Specific examples

[0964] 1. Coordination suggestions at apparel shops

[0965] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[0966] The terminal (robot) converts this voice into text data and sends it to the server.

[0967] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[0968] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[0969] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it asks a question aloud, "What do you think of this outfit?", while displaying an image of a specific outfit on the screen.

[0970] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[0971] 2. Menu suggestions at restaurants

[0972] The user asks verbally, "What are your recommendations for today?"

[0973] The terminal (robot) converts this voice into text data and sends it to the server.

[0974] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[0975] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[0976] The server will then use the emotion data to refine suggestions for celebrations or special occasions. For example, it might suggest, "Today is a special day, so how about this special menu?" while showing visual images.

[0977] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[0978] The processing flow will be explained below.

[0979] Step 1:

[0980] The user types a question or request in voice or text form, for example, "What should I wear today?"

[0981] Step 2:

[0982] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[0983] Step 3:

[0984] The device sends the converted text data to the server, and also sends the audio file if necessary.

[0985] Step 4:

[0986] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[0987] Step 5:

[0988] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[0989] Step 6:

[0990] The device (robot) captures the user's facial expressions and voice tone with a camera and microphone and applies emotion recognition algorithms to determine, for example, whether the user is nervous or relaxed.

[0991] Step 7:

[0992] The device generates emotion data and transmits it to the server. For example, it transmits data indicating that the user is nervous.

[0993] Step 8:

[0994] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[0995] Step 9:

[0996] The server transmits the generated proposal and response data to the terminal, for example, image data of a specific outfit.

[0997] Step 10:

[0998] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[0999] Step 11:

[1000] The user can review the suggestions and ask further questions or make requests, for example, "Are there any more formal options?"

[1001] Step 12:

[1002] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[1003] Example 1

[1004] 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."

[1005] Conventional dialogue systems simply analyze user input, making it difficult to generate responses that take the user's emotions and context into account. Furthermore, the low accuracy of voice input analysis makes it impossible to achieve a dialogue that satisfies the user. Therefore, there is a demand for a system that can have natural dialogue with users and provide optimal suggestions.

[1006] 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.

[1007] In this invention, the server includes means for accepting user input in voice or text format, means for converting the voice input into text data, means including a generative AI model for analyzing the user input, means for analyzing the user's facial expressions and tone of voice by applying an emotion recognition algorithm to generate emotion data, means for generating optimal suggestions and responses based on the user's input and emotion data, and means for providing the generated suggestions and responses to the user audio or visually. This makes it possible to provide natural dialogue and highly accurate suggestions that take the user's emotions and situation into consideration.

[1008] A "means for accepting user input in the form of voice or text" is an interface through which a user can make voice or text questions or requests to the system.

[1009] "Means for converting voice input into text data" refers to voice recognition technology for converting voice data received from a user into text data.

[1010] A "server containing a generative AI model for analyzing user input" is a server that has a model that utilizes machine learning to analyze a user's text data and generate appropriate responses or suggestions.

[1011] "Means of applying emotion recognition algorithms to analyze a user's facial expressions and tone of voice to generate emotion data" refers to algorithms that use a camera or microphone to capture a user's facial expressions and tone of voice and then analyze emotions based on that.

[1012] "Means for generating optimal suggestions and responses based on user input and emotional data" refers to the process of integrating the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[1013] "Means for providing generated suggestions or responses to the user via voice or visually" refers to means for delivering responses obtained from a generative AI model to the user using voice synthesis or a display.

[1014] This invention relates to a system that analyzes user input and generates optimal suggestions and responses by combining a generative AI model with an emotion recognition algorithm.

[1015] System configuration

[1016] 1. Accepting user input

[1017] The user asks a question or makes a request via voice or text, for example, "What kind of clothes would you like to wear today?"

[1018] The terminal (robot) accepts voice input and converts it into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). It can also accept text input directly.

[1019] 2. Data submission and analysis

[1020] The device sends the converted text data to the server using an API request.

[1021] The server inputs the received text data into a generative AI model (e.g., OpenAI GPT-4) and analyzes the question. At this time, the server generates a prompt and passes the data to the generative AI model. An example of a prompt is: "The user is asking about what to wear to dinner. Please give us the best suggestions."

[1022] 3. Emotion recognition

[1023] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, and uses emotion recognition algorithms (e.g., Hume AI) to process this data in real time.

[1024] The device sends the generated emotion data to the server. Specifically, it transmits information obtained as emotion data, such as "the user is nervous" or "relaxed."

[1025] 4. Generating the Response

[1026] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, for example, suggesting casual clothing if the user is relaxed.

[1027] The generated response data is saved in text format and sent to the terminal.

[1028] 5. User Feedback

[1029] The device converts the text response sent by the server into speech and provides it to the user, using text-to-speech software (e.g., Amazon Polly) to generate specific responses.

[1030] At the same time, images and text are displayed on the device's display to visually display the response. For example, an image of an outfit is displayed on the display along with the voice message, "What do you think of this outfit?"

[1031] Specific examples

[1032] 1. Coordination suggestions at apparel shops

[1033] User: "I'm going to dinner with a friend today. Can you tell me what to wear?"

[1034] Terminal: Converts voice into text data and sends it to the server.

[1035] Server: Analyzes questions using a generative AI model (e.g., GPT-4) and generates optimal outfits based on weather information and the latest fashion trends.

[1036] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[1037] Server: Corrects the suggestions based on emotional data, asks aloud "What do you think of this outfit?" as a concrete example, and displays an image on the screen.

[1038] User: Review the suggestions and make a follow-up request: "Do you have a more casual jacket?"

[1039] 2. Menu suggestions at restaurants

[1040] User: "What's your recommendation for today?"

[1041] Terminal: Converts voice into text data and sends it to the server.

[1042] Server: Analyzes the question using a generative AI model (e.g., GPT-4) and generates recommendations based on menu information and seasonal specials.

[1043] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[1044] Server: Corrects suggestions based on emotional data, suggests via voice, "Today is a special day, how about this special menu?" and shows a visual image.

[1045] User: Review the suggestions and ask further questions or make requests.

[1046] The system can engage in natural dialogue with users and provide appropriate and personalized suggestions and responses. Specific operations include voice recognition, data transmission and analysis, emotion recognition, response generation, speech synthesis, and display.

[1047] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1048] Step 1:

[1049] The user asks a question or makes a request in the form of voice or text. In this case, the input from the user is voice data or text data. For example, a voice input might be "What's the weather like today?" The device receives this voice and converts the voice data into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). The converted text data is output.

[1050] Step 2:

[1051] The device sends the converted text data to the server. The sent text data is securely sent to the server via an API request. The server uses a generative AI model (e.g., OpenAI GPT-4) to analyze the received text data. At this point, the server generates a prompt and passes the prompt and text data to the generative AI model. An example of a prompt is: "The user is asking about the weather. Please provide the latest weather information." The analysis result is output.

[1052] Step 3:

[1053] The device uses a camera and microphone to capture the user's facial expressions and voice tone. This captured data is analyzed using emotion recognition algorithms (e.g., Hume AI) to generate emotional data, such as whether the user is nervous or relaxed. The emotional data is output.

[1054] Step 4:

[1055] The device sends the generated emotional data to the server. This emotional data is then integrated with the text data analyzed earlier. The server generates optimal suggestions and responses based on the analysis results of the generative AI model and the emotional data. For example, it adjusts the content to provide weather information in softer language if the user is relaxed. The optimal suggestions and responses are then output.

[1056] Step 5:

[1057] The server saves the generated suggestions and responses as text data and sends this data to the device. The device converts this text response data into speech using speech synthesis software (e.g., Amazon Polly) and provides it to the user. The device also displays the response on the display. Specifically, it displays an image of a sunny day along with a voice such as "Today's weather is sunny." Feedback to the user is then complete.

[1058] (Application example 1)

[1059] 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."

[1060] Conventional user interface systems simply provide information in response to user input, making it difficult to make flexible suggestions based on the user's emotions and situation. In particular, when dealing with customers in physical stores, there is a demand for technology that can make appropriate suggestions and guidance based on emotional information such as the user's facial expression and tone of voice, but no system that can achieve this exists.

[1061] 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.

[1062] In this invention, the server includes means for accepting user input in voice or text format, means for cooperating with the server including a generative AI model for analyzing the user input, means for recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the user's input and emotion data, means for providing the generated suggestions and responses to the user in voice or visual form, and means for correcting the content of suggestions based on the user's emotions. This makes it possible to understand the user's emotions and provide personalized suggestions and guidance that are more closely aligned with the user's needs, particularly in physical stores.

[1063] "User Input" means any question or request made by a User to the System in the form of speech or text.

[1064] A "generative AI model" is an artificial intelligence model that can understand and analyze data such as text, audio, images, and video using machine learning algorithms.

[1065] An "emotion recognition algorithm" is an algorithm that analyzes a user's facial expressions and tone of voice to generate emotional data for the user.

[1066] "Suggestion and response generation" is the process by which the system creates optimal suggestions and responses based on the user's input and emotional data.

[1067] "Audio or visual presentation means" means a means of conveying generated suggestions or responses to the user audibly or visually displaying them on a device such as a display screen or smart mirror.

[1068] "Emotion-based suggestion adjustment" is the process of taking into account the user's emotional data and adjusting the generated suggestions and responses to suit the user's situation and mood.

[1069] This invention is a system that analyzes user input in brick-and-mortar stores, recognizes emotions, and provides appropriate suggestions and responses. Specifically, it uses a system that combines a generative AI model and an emotion recognition algorithm to generate optimal suggestions and responses based on the user's voice or text input and emotional data, and provides them to the user.

[1070] System configuration

[1071] 1. Accepting user input

[1072] Users can ask questions or make requests via voice or text. Smart mirrors and terminals installed in physical stores accept voice input and convert it into text data, and can also accept text input directly.

[1073] 2. Data submission and analysis

[1074] The smart mirror or device sends text data to a server, which then inputs the received text data into a generative AI model to analyze the question. The generative AI model understands text, audio, images, and video and generates appropriate responses and suggestions.

[1075] 3. Emotion recognition

[1076] The smart mirror or device captures the user's facial expressions and voice tone through a camera and microphone, then applies emotion recognition algorithms to the device, generating emotion data that is then sent to a server.

[1077] 4. Generating the Response

[1078] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, and further adjusts the suggestions based on the user's emotions.

[1079] 5. User Feedback

[1080] The smart mirror or device will generate suggestions and responses and provide them to the user both audibly and visually, for example by providing audio instructions accompanied by diagrams on the display.

[1081] Hardware and software used

[1082] Hardware:

[1083] Microphone: Captures the user's voice input.

[1084] Camera: Captures the user's facial expressions.

[1085] Smart mirror: Capable of displaying images and outputting audio.

[1086] software:

[1087] OpenAI API: Used to understand user questions and generate appropriate suggestions and responses through generative AI models.

[1088] SpeechRecognition: Used to convert voice input into text data.

[1089] EmotionRecognition: The algorithm used to recognize emotions from the user's face.

[1090] Pillow: A library for image processing.

[1091] Specific examples

[1092] For example, consider a case where a user speaks to a smart mirror installed in an apparel shop and asks, "I have dinner plans today, so what clothes should I wear?" The smart mirror converts this speech into text and inputs it into a generative AI model. The generative AI model generates appropriate fashion advice based on the input question. At the same time, the smart mirror captures the user's facial expressions with a camera and generates emotion data using an emotion recognition algorithm.

[1093] The server then adjusts the recommendations to best suit the user's situation based on the fashion advice and emotional data generated by the generative AI model. For example, if the server recognizes that the user is relaxed, it generates a response such as, "How about this chic dress? We also have a more casual option for a more relaxed look."

[1094] The generated suggestions are provided to the user via voice and display. For example, the following prompts are input to the generative AI model:

[1095] Today's user asked: 'I have a dinner appointment tonight, what suitable clothes do you recommend?' Provide a fashion advice considering user's preferences.

[1096] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1097] Step 1:

[1098] Accepting user input

[1099] The user asks a question or makes a request by voice or text. The device (such as a smart mirror) accepts the voice input and converts it into text data. It can also accept text input directly.

[1100] Input: User voice or text input

[1101] Output: Text data

[1102] Specific operation: Captures audio with a microphone and converts it to text using the SpeechRecognition library. Text input is accepted as text data.

[1103] Step 2:

[1104] Sending data

[1105] The terminal transmits the converted text data to the server.

[1106] Input: Text data

[1107] Output: Text data sent to the server

[1108] Specific operation: Use the device's communication function to send the converted text data to the server.

[1109] Step 3:

[1110] Data analysis

[1111] The server inputs the received text data into a generative AI model to analyze the user's question, which then generates appropriate responses and suggestions based on the analysis results.

[1112] Input: Text data sent to the server

[1113] Output: Analysis results from the generative AI model

[1114] Specific operation: Using the OpenAI API, text data is input into the generative AI model, a prompt sentence is created, and analysis results are obtained.

[1115] Step 4:

[1116] emotion recognition

[1117] The device captures the user's facial expressions and voice tone through a camera and microphone, applies emotion recognition algorithms to generate emotion data, and then transmits the emotion data to a server.

[1118] Input: User facial and voice data

[1119] Output: Emotion data

[1120] Specific operations: Capture the user's facial expressions with the camera, apply the EmotionRecognition algorithm to analyze the emotions and generate emotion data, capture the voice tone with the microphone, and apply the emotion recognition algorithm.

[1121] Step 5:

[1122] Generating a response

[1123] The server generates optimal suggestions and responses based on the analysis results of the generative AI model and emotional data, and also adjusts the suggestions based on the user's emotions.

[1124] Input: Analysis results of generative AI model, emotion data

[1125] Output: Best suggestion or response

[1126] Specific operation: Emotional data is integrated into the analysis results obtained by the generative AI model to generate personalized suggestions and responses that take into account the user's emotional state.

[1127] Step 6:

[1128] User feedback

[1129] The final suggestion or response is provided to the user via audio and visual means by the device (e.g., smart mirror).

[1130] Input: Best suggestion or response

[1131] Output: Audio and visual feedback provided to the user

[1132] Specific operation: Suggestions and responses are displayed on the smart mirror's display and communicated to the user using voice output.

[1133] 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.

[1134] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[1135] System configuration

[1136] 1. Accepting user input

[1137] The user asks a question or requests via voice or text, for example, "What should I wear today?"

[1138] The terminal (robot) accepts voice input and converts it into text data. If the input is text, it is accepted as is.

[1139] 2. Data submission and analysis

[1140] The terminal (robot) sends the converted text data to the server.

[1141] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[1142] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1143] 3. Emotion recognition

[1144] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1145] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[1146] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[1147] 4. Generating the Response

[1148] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1149] 5. User Feedback

[1150] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[1151] Specific examples

[1152] 1. Coordination suggestions at apparel shops

[1153] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[1154] The terminal (robot) converts this voice into text data and sends it to the server.

[1155] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[1156] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[1157] The emotion engine analyzes this emotion information to generate emotion data, which is then sent to the server.

[1158] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it might ask a question aloud, such as, "How about this relaxed, casual outfit?", while showing specific outfit images on the display.

[1159] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[1160] 2. Menu suggestions at restaurants

[1161] The user asks verbally, "What are your recommendations for today?"

[1162] The terminal (robot) converts this voice into text data and sends it to the server.

[1163] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[1164] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[1165] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. For example, it suggests "Today is a special day, so how about this special menu?" while showing a visual image.

[1166] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[1167] The processing flow will be explained below.

[1168] Step 1:

[1169] The user types a question or request in voice or text form, for example, "What should I wear today?"

[1170] Step 2:

[1171] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[1172] Step 3:

[1173] The terminal (robot) sends the converted text data to the server. In the case of voice input, the audio file is also sent.

[1174] Step 4:

[1175] The server inputs the received text data into a generative AI model and analyzes the content of the user's question. For example, it identifies the intent, such as "I'm looking for clothing suggestions."

[1176] Step 5:

[1177] The generative AI model understands text, speech, images, and video, and gathers information from relevant databases and external APIs, such as the latest fashion trends and weather data, to generate optimal outfits.

[1178] Step 6:

[1179] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, thereby understanding the user's emotional state.

[1180] Step 7:

[1181] The device sends the captured facial and voice data to an emotion engine, which runs emotion recognition algorithms, for example, to analyze whether the user is nervous or relaxed.

[1182] Step 8:

[1183] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it sends data such as "The user is nervous."

[1184] Step 9:

[1185] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting relaxing, casual clothing.

[1186] Step 10:

[1187] The server transmits the generated proposal and response data to the terminal, for example, data including a specific coordinated image and description.

[1188] Step 11:

[1189] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it may ask a question aloud, such as "How about this casual jacket?", while showing an image of a specific outfit on the display.

[1190] Step 12:

[1191] The user can review the suggestions and ask further questions or make requests. For example, "Are there any more formal options?"

[1192] Step 13:

[1193] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[1194] In this way, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses based on emotion recognition.

[1195] Example 2

[1196] 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."

[1197] Conventional dialogue systems have difficulty fully understanding the user's intentions and emotions and providing appropriate responses and suggestions. Furthermore, advanced processing is required to convert voice input into text and analyze user input in real time, but systems that integrate these processes remain limited. Therefore, there is a need for a system that can accurately analyze user input and provide responses that take emotions into account.

[1198] 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.

[1199] In this invention, the server includes means for accepting user input in the form of voice or text, means for using voice recognition software to convert the accepted voice data into text data, means for transmitting the converted text data to the server, means for analyzing the user input in cooperation with the server including the generative AI model, means for capturing the user's facial expressions and tone of voice using a camera and a microphone, means for analyzing and recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the analysis results of the generative AI model and the emotion data, and means for providing the generated suggestions and responses to the user audio or visually, thereby enabling accurate analysis of the user's input and providing personalized responses and suggestions that take emotions into consideration.

[1200] "Voice or textual input means" refers to an interface for receiving voice or textual input from a user and processing that data.

[1201] "Speech recognition software" refers to a program used to convert voice data into text data. A specific example is automatic speech recognition (ASR) technology.

[1202] A "server" refers to a computer system connected to a network for processing requests from multiple clients.

[1203] A "generative AI model" refers to an artificial intelligence algorithm that analyzes user input data and generates appropriate responses or suggestions based on that data. A specific example is a natural language processing model.

[1204] "Means of capturing using a camera and microphone" refers to a method of using a camera and microphone to capture a user's facial expressions and tone of voice.

[1205] "Emotion recognition algorithm" refers to an algorithm that analyzes acquired data (facial expressions and voice tone) to recognize the user's emotional state (e.g., joy, sadness, tension, etc.).

[1206] "Emotion engine" refers to a software or hardware configuration for executing emotion recognition algorithms.

[1207] "Means for generating suggestions or responses" refers to a method for creating appropriate suggestions or responses for a user based on the analysis results and emotional data.

[1208] "Audio or visual means" refers to a method for communicating generated suggestions or responses to the user, such as by audio output or display.

[1209] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. The system receives user input in the form of voice or text, analyzes the input, recognizes the user's emotion, and then provides appropriate responses and suggestions.

[1210] System configuration

[1211] 1. Accepting user input

[1212] Users can enter questions or requests in voice or text format, for example, "What should I wear today?"

[1213] The terminal (robot) receives this voice and converts it into text data using voice recognition software (for example, a voice recognition API). If the input is in text format, it is accepted by the system as is.

[1214] 2. Data submission and analysis

[1215] The terminal transmits the converted text data to the server.

[1216] The server inputs the received text data into a generative AI model (specifically, a natural language processing model) and analyzes the question. For example, it identifies the theme "wearing clothes."

[1217] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data APIs, fashion database APIs). For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1218] 3. Emotion recognition

[1219] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1220] The device sends the acquired data to an emotion engine, which applies emotion recognition algorithms (e.g., machine learning algorithms) to determine, for example, whether the user is tense or relaxed.

[1221] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates information such as "the user is nervous."

[1222] 4. Generating the Response

[1223] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1224] 5. User Feedback

[1225] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example, saying "How about this casual jacket?" and showing an image on the display.

[1226] Specific examples

[1227] Coordination suggestions at apparel shops

[1228] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[1229] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[1230] The server inputs the text data into a generative AI model (natural language processing model) to analyze the user's intentions, and then generates optimal outfits based on weather information and the latest fashion trends (fashion database API).

[1231] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional information such as whether the user is nervous or relaxed.

[1232] The emotion engine analyzes the emotion information to generate emotion data, which is then sent to the server.

[1233] Based on the acquired emotional data, the server adjusts its suggestions to match the user's preferences, asking aloud, "How about a relaxed, casual outfit?" while displaying specific outfit images on the screen.

[1234] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[1235] Menu suggestions at restaurants

[1236] The user asks verbally, "What are your recommendations for today?"

[1237] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[1238] The server inputs the text data into a generative AI model (natural language processing model) and analyzes the question. It then generates recommended menu items by referencing menu information and seasonal special dishes (special dish database).

[1239] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional data.

[1240] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. It suggests aloud, "Today is a special day, so how about this special menu?" while showing visual images.

[1241] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[1242] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1243] Step 1: Accept user input

[1244] The user types in a question or request through voice or text, for example, "What should I wear today?"

[1245] Input: User voice or text data

[1246] The terminal (robot) receives voice using a microphone and converts the voice into text data using voice recognition software (e.g., voice recognition API). If the input is text, it is accepted by the system as is.

[1247] Output: Converted text data

[1248] Specific behavior:

[1249] The device collects the user's voice through a microphone.

[1250] Call the speech recognition API and convert the speech data into text data.

[1251] The converted text data is temporarily stored.

[1252] Step 2: Sending data

[1253] The device then sends the converted text data to the server.

[1254] Input: Text data stored on the device

[1255] Output: Text data sent to the server

[1256] Specific behavior:

[1257] The converted text data is sent to the server via the network.

[1258] Verify that the submission was successful.

[1259] Step 3: Analyze the data

[1260] The server inputs the received text data into a generative AI model (e.g., a natural language processing model) and analyzes the question content.

[1261] Input: Text data

[1262] Output: Parsed question topic (e.g. "Outfit suggestions")

[1263] Specific behavior:

[1264] The server preprocesses the received text data and generates a prompt (e.g., "The user is asking, 'What should I wear today?' Please generate an appropriate response.").

[1265] The prompt sentence is input into the generative AI model, and the model begins analysis.

[1266] As a result of the analysis, the topic of the question and related information are identified.

[1267] Step 4: Gather information

[1268] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data API, fashion database API).

[1269] Input: Question topic, related information retrieved from databases and external APIs

[1270] Output: Analysis results (e.g., optimal outfits based on the latest fashion trends or weather data)

[1271] Specific behavior:

[1272] The generative AI model calls database APIs or external APIs to obtain the necessary information.

[1273] Based on the acquired information, the model performs analysis and generates coordination suggestions.

[1274] Step 5: Emotion Recognition

[1275] The device uses a camera and microphone to capture the user's facial expressions and voice tone.

[1276] Input: Camera video data, audio tone data

[1277] Output: Captured facial expression data and voice tone data

[1278] Specific behavior:

[1279] The device captures the user's facial expressions with a camera and collects voice tones with a microphone.

[1280] The collected facial expression data and voice tone data are temporarily stored.

[1281] Step 6: Analyze the sentiment data

[1282] The device sends the acquired data to the emotion engine and applies an emotion recognition algorithm (e.g., a machine learning algorithm).

[1283] Input: facial expression data, voice tone data

[1284] Output: Emotion data (e.g., "The user is nervous")

[1285] Specific behavior:

[1286] Facial expression data and voice tone data are sent to the emotion engine.

[1287] An emotion recognition algorithm is used to analyze emotions and generate emotion data.

[1288] The generated emotion data is sent to the server.

[1289] Step 7: Generate a response

[1290] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[1291] Input: Analysis results (e.g., outfit suggestions), emotion data

[1292] Output: Optimal suggestion or response (e.g., "Suggestions for relaxing casual clothing")

[1293] Specific behavior:

[1294] Integrate analysis results with emotion data.

[1295] It takes into account the user's emotional state to instruct the generative AI model on how to generate a response.

[1296] Generate optimal proposals and responses and send them to the device.

[1297] Step 8: User feedback

[1298] The terminal (robot) provides the generated suggestions and responses to the user audibly and visually.

[1299] Input: Best suggestion or response

[1300] Output: Audio and visual suggestions and responses provided to the user

[1301] Specific behavior:

[1302] The terminal uses speech synthesis software to audibly output the generated suggestions.

[1303] At the same time, an image of the coordinated outfit is displayed on the display.

[1304] The user can review the information provided and ask further questions or requests, for example, "Do you have a more casual jacket?"

[1305] (Application example 2)

[1306] 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."

[1307] In today's brick-and-mortar stores, it is not easy for users to receive accurate and personalized suggestions when selecting products or using services. Conventional systems have difficulty not only analyzing user input, but also properly recognizing the user's emotional state and making optimal suggestions based on this. This has resulted in the problem of insufficient effective customer service and service provision.

[1308] 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 accepting user input in voice or text format; means including a generative AI model for analyzing the user input; means for recognizing the user's emotion by applying an emotion recognition algorithm; means for generating optimal suggestions and responses based on the user's input and emotion data; means for providing the generated suggestions and responses to the user audio or visually; and means for analyzing the user's facial expressions and tone of voice using a camera and microphone when the user asks a question in a physical store, and correcting the generative AI model based on the emotion data to make optimal suggestions. This enables personalized suggestions and responses based on the user's emotional state.

[1309] "User Input" means a question or request provided by a User in voice or text form.

[1310] A "generative AI model" is an artificial intelligence model that analyzes user input, understands text, voice, images, and video, and generates optimal suggestions and responses.

[1311] The "emotion recognition algorithm" is an algorithm that uses a camera and microphone to analyze a user's facial expressions and tone of voice to understand their emotional state.

[1312] "Emotional data" is data that indicates a user's emotional state as analyzed by an emotion recognition algorithm.

[1313] "Suggestions and responses" refer to optimal information and suggestions derived by the generative AI model based on user input and emotional data.

[1314] "Means for providing audio or visual information to the user" refers to means for conveying generated suggestions or responses to the user through audio output or visual information such as a display.

[1315] A "brick and mortar store" is a physical location where consumers can visit and consume products or services.

[1316] "Camera and microphone" refers to video and audio input devices for capturing the user's facial expressions and voice tone.

[1317] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[1318] System configuration

[1319] 1. Accepting user input

[1320] The device provides an interface through which the user can ask questions or make requests in the form of voice or text, for example, by saying, "What should I wear today?"

[1321] The device accepts voice input and converts it into text data, or accepts text input as is.

[1322] 2. Data submission and analysis

[1323] The terminal transmits the converted text data to the server.

[1324] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[1325] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1326] 3. Emotion recognition

[1327] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1328] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[1329] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[1330] 4. Generating the Response

[1331] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1332] 5. User Feedback

[1333] The device will then provide the generated suggestions and responses to the user both audibly and visually, for example, asking a question aloud, "What do you think of this casual jacket?", while showing an image on the display.

[1334] Hardware and software used

[1335] 1. Hardware

[1336] camera

[1337] microphone

[1338] Smart glasses, head-mounted displays, tablets

[1339] 2. Software

[1340] Emotion recognition libraries (e.g., Affectiva)

[1341] Generative AI models (e.g., OpenAI GPT-4)

[1342] Speech processing engine (e.g. Google Speech-to-Text)

[1343] GUI library for display (e.g. Qt)

[1344] Examples of concrete examples and prompts

[1345] Examples:

[1346] When a user asks "What's your recommended outfit for today?" and the emotion recognition system determines that the answer is "Relaxed," the prompt is:

[1347] User Question: What's your go-to outfit for today?

[1348] The user is relaxed.

[1349] Please make appropriate suggestions.

[1350] Thus, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses.

[1351] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1352] Step 1:

[1353] A user inputs a question or request to a terminal in the form of voice or text. This is done through a voice interface or a text input interface. For example, the question "What should I wear today?" is input.

[1354] Input: User voice or text question

[1355] Output: Audio data or text data

[1356] Step 2:

[1357] The device converts the voice input into text data. If the input is voice, a speech processing engine (e.g., Google Speech-to-Text) converts the voice into text. If the input is text, this step is skipped.

[1358] Input: Audio data

[1359] Output: Text data

[1360] Step 3:

[1361] The device sends text data to the server, which receives it and inputs it into the generative AI model.

[1362] Input: Text data

[1363] Output: Request for analysis results

[1364] Step 4:

[1365] The server analyzes the text data and understands the question using a generative AI model (e.g., OpenAI GPT-4). The generative AI model collects and analyzes relevant information to generate optimal suggestions, taking into account, for example, the latest fashion trends and weather data.

[1366] Input: Text data

[1367] Output: Initial proposal data

[1368] Step 5:

[1369] The device uses a camera and microphone to capture the user's facial expressions and voice tone, and this data is sent to an emotion recognition algorithm (e.g., Affectiva) to analyze the user's emotional state.

[1370] Input: User's face image and voice tone

[1371] Output: Emotion data

[1372] Step 6:

[1373] The emotion engine generates emotion data analyzed by an emotion recognition algorithm and sends it to the server, which then provides emotion information such as "the user is nervous."

[1374] Input: Facial image data and voice tone data

[1375] Output: Parsed emotion data

[1376] Step 7:

[1377] The server combines the analysis results of the generated AI model with emotional data to generate optimal suggestions and responses. For example, it may suggest "casual clothing that is comfortable to wear."

[1378] Input: Initial proposal data and emotion data

[1379] Output: Optimized proposal data

[1380] Step 8:

[1381] The device provides the generated suggestions and responses to the user both audibly and visually. Using a voice output engine and a display, the device displays an image on the screen while providing a voice prompt such as, "How about this casual jacket?"

[1382] Input: Optimized proposal data

[1383] Output: Audio audio output and visual display

[1384] In this way, each step processes specific input data and produces an appropriate output based on that data, allowing for personalized suggestions and responses to be provided to the user.

[1385] 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.

[1386] 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.

[1387] 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.

[1388] [Fourth embodiment]

[1389] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1390] 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.

[1391] 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).

[1392] 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.

[1393] 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.

[1394] 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).

[1395] 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.

[1396] 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.

[1397] 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.

[1398] 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.

[1399] 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.

[1400] 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.

[1401] 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."

[1402] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion-recognition algorithm. Specifically, this system is implemented as follows:

[1403] System configuration

[1404] 1. Accepting user input

[1405] The user asks a question or makes a request via voice or text.

[1406] The terminal (robot) accepts voice input and converts it into text data. It also accepts text input directly.

[1407] 2. Data submission and analysis

[1408] The terminal (robot) sends text data to the server.

[1409] The server inputs the received text data into a generative AI model (a model using a machine learning algorithm) and analyzes the question content.

[1410] Generative AI models understand text, speech, images, and video to generate appropriate responses and suggestions.

[1411] 3. Emotion recognition

[1412] The device (robot) captures the user's facial expressions and tone of voice through a camera and microphone and applies emotion recognition algorithms.

[1413] The terminal generates emotion data and transmits it to the server.

[1414] 4. Generating the Response

[1415] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data.

[1416] 5. User Feedback

[1417] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example by providing an explanation audibly while showing an illustration on the display.

[1418] Specific examples

[1419] 1. Coordination suggestions at apparel shops

[1420] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[1421] The terminal (robot) converts this voice into text data and sends it to the server.

[1422] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[1423] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[1424] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it asks a question aloud, "What do you think of this outfit?", while displaying an image of a specific outfit on the screen.

[1425] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[1426] 2. Menu suggestions at restaurants

[1427] The user asks verbally, "What are your recommendations for today?"

[1428] The terminal (robot) converts this voice into text data and sends it to the server.

[1429] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[1430] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[1431] The server will then use the emotion data to refine suggestions for celebrations or special occasions. For example, it might suggest, "Today is a special day, so how about this special menu?" while showing visual images.

[1432] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[1433] The processing flow will be explained below.

[1434] Step 1:

[1435] The user types a question or request in voice or text form, for example, "What should I wear today?"

[1436] Step 2:

[1437] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[1438] Step 3:

[1439] The device sends the converted text data to the server, and also sends the audio file if necessary.

[1440] Step 4:

[1441] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[1442] Step 5:

[1443] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1444] Step 6:

[1445] The device (robot) captures the user's facial expressions and voice tone with a camera and microphone and applies emotion recognition algorithms to determine, for example, whether the user is nervous or relaxed.

[1446] Step 7:

[1447] The device generates emotion data and transmits it to the server. For example, it transmits data indicating that the user is nervous.

[1448] Step 8:

[1449] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1450] Step 9:

[1451] The server transmits the generated proposal and response data to the terminal, for example, image data of a specific outfit.

[1452] Step 10:

[1453] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[1454] Step 11:

[1455] The user can review the suggestions and ask further questions or make requests, for example, "Are there any more formal options?"

[1456] Step 12:

[1457] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[1458] Example 1

[1459] 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."

[1460] Conventional dialogue systems simply analyze user input, making it difficult to generate responses that take the user's emotions and context into account. Furthermore, the low accuracy of voice input analysis makes it impossible to achieve a dialogue that satisfies the user. Therefore, there is a demand for a system that can have natural dialogue with users and provide optimal suggestions.

[1461] 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.

[1462] In this invention, the server includes means for accepting user input in voice or text format, means for converting the voice input into text data, means including a generative AI model for analyzing the user input, means for analyzing the user's facial expressions and tone of voice by applying an emotion recognition algorithm to generate emotion data, means for generating optimal suggestions and responses based on the user's input and emotion data, and means for providing the generated suggestions and responses to the user audio or visually. This makes it possible to provide natural dialogue and highly accurate suggestions that take the user's emotions and situation into consideration.

[1463] A "means for accepting user input in the form of voice or text" is an interface through which a user can make voice or text questions or requests to the system.

[1464] "Means for converting voice input into text data" refers to voice recognition technology for converting voice data received from a user into text data.

[1465] A "server containing a generative AI model for analyzing user input" is a server that has a model that utilizes machine learning to analyze a user's text data and generate appropriate responses or suggestions.

[1466] "Means of applying emotion recognition algorithms to analyze a user's facial expressions and tone of voice to generate emotion data" refers to algorithms that use a camera or microphone to capture a user's facial expressions and tone of voice and then analyze emotions based on that.

[1467] "Means for generating optimal suggestions and responses based on user input and emotional data" refers to the process of integrating the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[1468] "Means for providing generated suggestions or responses to the user via voice or visually" refers to means for delivering responses obtained from a generative AI model to the user using voice synthesis or a display.

[1469] This invention relates to a system that analyzes user input and generates optimal suggestions and responses by combining a generative AI model with an emotion recognition algorithm.

[1470] System configuration

[1471] 1. Accepting user input

[1472] The user asks a question or makes a request via voice or text, for example, "What kind of clothes would you like to wear today?"

[1473] The terminal (robot) accepts voice input and converts it into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). It can also accept text input directly.

[1474] 2. Data submission and analysis

[1475] The device sends the converted text data to the server using an API request.

[1476] The server inputs the received text data into a generative AI model (e.g., OpenAI GPT-4) and analyzes the question. At this time, the server generates a prompt and passes the data to the generative AI model. An example of a prompt is: "The user is asking about what to wear to dinner. Please give us the best suggestions."

[1477] 3. Emotion recognition

[1478] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, and uses emotion recognition algorithms (e.g., Hume AI) to process this data in real time.

[1479] The device sends the generated emotion data to the server. Specifically, it transmits information obtained as emotion data, such as "the user is nervous" or "relaxed."

[1480] 4. Generating the Response

[1481] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, for example, suggesting casual clothing if the user is relaxed.

[1482] The generated response data is saved in text format and sent to the terminal.

[1483] 5. User Feedback

[1484] The device converts the text response sent by the server into speech and provides it to the user, using text-to-speech software (e.g., Amazon Polly) to generate specific responses.

[1485] At the same time, images and text are displayed on the device's display to visually display the response. For example, an image of an outfit is displayed on the display along with the voice message, "What do you think of this outfit?"

[1486] Specific examples

[1487] 1. Coordination suggestions at apparel shops

[1488] User: "I'm going to dinner with a friend today. Can you tell me what to wear?"

[1489] Terminal: Converts voice into text data and sends it to the server.

[1490] Server: Analyzes questions using a generative AI model (e.g., GPT-4) and generates optimal outfits based on weather information and the latest fashion trends.

[1491] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[1492] Server: Corrects the suggestions based on emotional data, asks aloud "What do you think of this outfit?" as a concrete example, and displays an image on the screen.

[1493] User: Review the suggestions and make a follow-up request: "Do you have a more casual jacket?"

[1494] 2. Menu suggestions at restaurants

[1495] User: "What's your recommendation for today?"

[1496] Terminal: Converts voice into text data and sends it to the server.

[1497] Server: Analyzes the question using a generative AI model (e.g., GPT-4) and generates recommendations based on menu information and seasonal specials.

[1498] Device: Analyzes facial expressions and voice tones using emotion recognition algorithms (e.g., Hume AI) to obtain emotional data.

[1499] Server: Corrects suggestions based on emotional data, suggests via voice, "Today is a special day, how about this special menu?" and shows a visual image.

[1500] User: Review the suggestions and ask further questions or make requests.

[1501] The system can engage in natural dialogue with users and provide appropriate and personalized suggestions and responses. Specific operations include voice recognition, data transmission and analysis, emotion recognition, response generation, speech synthesis, and display.

[1502] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1503] Step 1:

[1504] The user asks a question or makes a request in the form of voice or text. In this case, the input from the user is voice data or text data. For example, a voice input might be "What's the weather like today?" The device receives this voice and converts the voice data into text data using voice recognition software (e.g., Google Cloud Speech-to-Text). The converted text data is output.

[1505] Step 2:

[1506] The device sends the converted text data to the server. The sent text data is securely sent to the server via an API request. The server uses a generative AI model (e.g., OpenAI GPT-4) to analyze the received text data. At this point, the server generates a prompt and passes the prompt and text data to the generative AI model. An example of a prompt is: "The user is asking about the weather. Please provide the latest weather information." The analysis result is output.

[1507] Step 3:

[1508] The device uses a camera and microphone to capture the user's facial expressions and voice tone. This captured data is analyzed using emotion recognition algorithms (e.g., Hume AI) to generate emotional data, such as whether the user is nervous or relaxed. The emotional data is output.

[1509] Step 4:

[1510] The device sends the generated emotional data to the server. This emotional data is then integrated with the text data analyzed earlier. The server generates optimal suggestions and responses based on the analysis results of the generative AI model and the emotional data. For example, it adjusts the content to provide weather information in softer language if the user is relaxed. The optimal suggestions and responses are then output.

[1511] Step 5:

[1512] The server saves the generated suggestions and responses as text data and sends this data to the device. The device converts this text response data into speech using speech synthesis software (e.g., Amazon Polly) and provides it to the user. The device also displays the response on the display. Specifically, it displays an image of a sunny day along with a voice such as "Today's weather is sunny." Feedback to the user is then complete.

[1513] (Application example 1)

[1514] 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."

[1515] Conventional user interface systems simply provide information in response to user input, making it difficult to make flexible suggestions based on the user's emotions and situation. In particular, when dealing with customers in physical stores, there is a demand for technology that can make appropriate suggestions and guidance based on emotional information such as the user's facial expression and tone of voice, but no system that can achieve this exists.

[1516] 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.

[1517] In this invention, the server includes means for accepting user input in voice or text format, means for cooperating with the server including a generative AI model for analyzing the user input, means for recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the user's input and emotion data, means for providing the generated suggestions and responses to the user in voice or visual form, and means for correcting the content of suggestions based on the user's emotions. This makes it possible to understand the user's emotions and provide personalized suggestions and guidance that are more closely aligned with the user's needs, particularly in physical stores.

[1518] "User Input" means any question or request made by a User to the System in the form of speech or text.

[1519] A "generative AI model" is an artificial intelligence model that can understand and analyze data such as text, audio, images, and video using machine learning algorithms.

[1520] An "emotion recognition algorithm" is an algorithm that analyzes a user's facial expressions and tone of voice to generate emotional data for the user.

[1521] "Suggestion and response generation" is the process by which the system creates optimal suggestions and responses based on the user's input and emotional data.

[1522] "Audio or visual presentation means" means a means of conveying generated suggestions or responses to the user audibly or visually displaying them on a device such as a display screen or smart mirror.

[1523] "Emotion-based suggestion adjustment" is the process of taking into account the user's emotional data and adjusting the generated suggestions and responses to suit the user's situation and mood.

[1524] This invention is a system that analyzes user input in brick-and-mortar stores, recognizes emotions, and provides appropriate suggestions and responses. Specifically, it uses a system that combines a generative AI model and an emotion recognition algorithm to generate optimal suggestions and responses based on the user's voice or text input and emotional data, and provides them to the user.

[1525] System configuration

[1526] 1. Accepting user input

[1527] Users can ask questions or make requests via voice or text. Smart mirrors and terminals installed in physical stores accept voice input and convert it into text data, and can also accept text input directly.

[1528] 2. Data submission and analysis

[1529] The smart mirror or device sends text data to a server, which then inputs the received text data into a generative AI model to analyze the question. The generative AI model understands text, audio, images, and video and generates appropriate responses and suggestions.

[1530] 3. Emotion recognition

[1531] The smart mirror or device captures the user's facial expressions and voice tone through a camera and microphone, then applies emotion recognition algorithms to the device, generating emotion data that is then sent to a server.

[1532] 4. Generating the Response

[1533] The server generates optimal suggestions and responses for the user based on the analysis results of the generative AI model and emotional data, and further adjusts the suggestions based on the user's emotions.

[1534] 5. User Feedback

[1535] The smart mirror or device will generate suggestions and responses and provide them to the user both audibly and visually, for example by providing audio instructions accompanied by diagrams on the display.

[1536] Hardware and software used

[1537] Hardware:

[1538] Microphone: Captures the user's voice input.

[1539] Camera: Captures the user's facial expressions.

[1540] Smart mirror: Capable of displaying images and outputting audio.

[1541] software:

[1542] OpenAI API: Used to understand user questions and generate appropriate suggestions and responses through generative AI models.

[1543] SpeechRecognition: Used to convert voice input into text data.

[1544] EmotionRecognition: The algorithm used to recognize emotions from the user's face.

[1545] Pillow: A library for image processing.

[1546] Specific examples

[1547] For example, consider a case where a user speaks to a smart mirror installed in an apparel shop and asks, "I have dinner plans today, so what clothes should I wear?" The smart mirror converts this speech into text and inputs it into a generative AI model. The generative AI model generates appropriate fashion advice based on the input question. At the same time, the smart mirror captures the user's facial expressions with a camera and generates emotion data using an emotion recognition algorithm.

[1548] The server then adjusts the recommendations to best suit the user's situation based on the fashion advice and emotional data generated by the generative AI model. For example, if the server recognizes that the user is relaxed, it generates a response such as, "How about this chic dress? We also have a more casual option for a more relaxed look."

[1549] The generated suggestions are provided to the user via voice and display. For example, the following prompts are input to the generative AI model:

[1550] Today's user asked: 'I have a dinner appointment tonight, what suitable clothes do you recommend?' Provide a fashion advice considering user's preferences.

[1551] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1552] Step 1:

[1553] Accepting user input

[1554] The user asks a question or makes a request by voice or text. The device (such as a smart mirror) accepts the voice input and converts it into text data. It can also accept text input directly.

[1555] Input: User voice or text input

[1556] Output: Text data

[1557] Specific operation: Captures audio with a microphone and converts it to text using the SpeechRecognition library. Text input is accepted as text data.

[1558] Step 2:

[1559] Sending data

[1560] The terminal transmits the converted text data to the server.

[1561] Input: Text data

[1562] Output: Text data sent to the server

[1563] Specific operation: Use the device's communication function to send the converted text data to the server.

[1564] Step 3:

[1565] Data analysis

[1566] The server inputs the received text data into a generative AI model to analyze the user's question, which then generates appropriate responses and suggestions based on the analysis results.

[1567] Input: Text data sent to the server

[1568] Output: Analysis results from the generative AI model

[1569] Specific operation: Using the OpenAI API, text data is input into the generative AI model, a prompt sentence is created, and analysis results are obtained.

[1570] Step 4:

[1571] emotion recognition

[1572] The device captures the user's facial expressions and voice tone through a camera and microphone, applies emotion recognition algorithms to generate emotion data, and then transmits the emotion data to a server.

[1573] Input: User facial and voice data

[1574] Output: Emotion data

[1575] Specific operations: Capture the user's facial expressions with the camera, apply the EmotionRecognition algorithm to analyze the emotions and generate emotion data, capture the voice tone with the microphone, and apply the emotion recognition algorithm.

[1576] Step 5:

[1577] Generating a response

[1578] The server generates optimal suggestions and responses based on the analysis results of the generative AI model and emotional data, and also adjusts the suggestions based on the user's emotions.

[1579] Input: Analysis results of generative AI model, emotion data

[1580] Output: Best suggestion or response

[1581] Specific operation: Emotional data is integrated into the analysis results obtained by the generative AI model to generate personalized suggestions and responses that take into account the user's emotional state.

[1582] Step 6:

[1583] User feedback

[1584] The final suggestion or response is provided to the user via audio and visual means by the device (e.g., smart mirror).

[1585] Input: Best suggestion or response

[1586] Output: Audio and visual feedback provided to the user

[1587] Specific operation: Suggestions and responses are displayed on the smart mirror's display and communicated to the user using voice output.

[1588] 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.

[1589] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[1590] System configuration

[1591] 1. Accepting user input

[1592] The user asks a question or requests via voice or text, for example, "What should I wear today?"

[1593] The terminal (robot) accepts voice input and converts it into text data. If the input is text, it is accepted as is.

[1594] 2. Data submission and analysis

[1595] The terminal (robot) sends the converted text data to the server.

[1596] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[1597] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1598] 3. Emotion recognition

[1599] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1600] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[1601] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[1602] 4. Generating the Response

[1603] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1604] 5. User Feedback

[1605] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it asks a question aloud, "How about this casual jacket?" while displaying an image on the display.

[1606] Specific examples

[1607] 1. Coordination suggestions at apparel shops

[1608] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[1609] The terminal (robot) converts this voice into text data and sends it to the server.

[1610] The server inputs the text data into a generative AI model, analyzes the user's intentions, and generates optimal outfits based on weather information and the latest fashion trends.

[1611] The device (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional information such as whether the user is nervous or relaxed.

[1612] The emotion engine analyzes this emotion information to generate emotion data, which is then sent to the server.

[1613] The server then uses the acquired emotional data to adjust its suggestions to match the user's preferences. For example, it might ask a question aloud, such as, "How about this relaxed, casual outfit?", while showing specific outfit images on the display.

[1614] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[1615] 2. Menu suggestions at restaurants

[1616] The user asks verbally, "What are your recommendations for today?"

[1617] The terminal (robot) converts this voice into text data and sends it to the server.

[1618] The server inputs the text data into a generative AI model, analyzes the question, and generates a menu recommendation based on menu information and seasonal specials.

[1619] The terminal (robot) analyzes the user's facial expressions and voice tone using an emotion recognition algorithm to obtain emotional data.

[1620] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. For example, it suggests "Today is a special day, so how about this special menu?" while showing a visual image.

[1621] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[1622] The processing flow will be explained below.

[1623] Step 1:

[1624] The user types a question or request in voice or text form, for example, "What should I wear today?"

[1625] Step 2:

[1626] The terminal (robot) receives voice input and converts it into text data. If the input is text, it is accepted as is.

[1627] Step 3:

[1628] The terminal (robot) sends the converted text data to the server. In the case of voice input, the audio file is also sent.

[1629] Step 4:

[1630] The server inputs the received text data into a generative AI model and analyzes the content of the user's question. For example, it identifies the intent, such as "I'm looking for clothing suggestions."

[1631] Step 5:

[1632] The generative AI model understands text, speech, images, and video, and gathers information from relevant databases and external APIs, such as the latest fashion trends and weather data, to generate optimal outfits.

[1633] Step 6:

[1634] The device (robot) uses a camera and microphone to capture the user's facial expressions and tone of voice, thereby understanding the user's emotional state.

[1635] Step 7:

[1636] The device sends the captured facial and voice data to an emotion engine, which runs emotion recognition algorithms, for example, to analyze whether the user is nervous or relaxed.

[1637] Step 8:

[1638] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it sends data such as "The user is nervous."

[1639] Step 9:

[1640] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting relaxing, casual clothing.

[1641] Step 10:

[1642] The server transmits the generated proposal and response data to the terminal, for example, data including a specific coordinated image and description.

[1643] Step 11:

[1644] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually. For example, it may ask a question aloud, such as "How about this casual jacket?", while showing an image of a specific outfit on the display.

[1645] Step 12:

[1646] The user can review the suggestions and ask further questions or make requests. For example, "Are there any more formal options?"

[1647] Step 13:

[1648] The terminal (robot) receives the new input, converts it back into text data, and sends it to the server, where the same process is repeated.

[1649] In this way, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses based on emotion recognition.

[1650] Example 2

[1651] 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."

[1652] Conventional dialogue systems have difficulty fully understanding the user's intentions and emotions and providing appropriate responses and suggestions. Furthermore, advanced processing is required to convert voice input into text and analyze user input in real time, but systems that integrate these processes remain limited. Therefore, there is a need for a system that can accurately analyze user input and provide responses that take emotions into account.

[1653] 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.

[1654] In this invention, the server includes means for accepting user input in the form of voice or text, means for using voice recognition software to convert the accepted voice data into text data, means for transmitting the converted text data to the server, means for analyzing the user input in cooperation with the server including the generative AI model, means for capturing the user's facial expressions and tone of voice using a camera and a microphone, means for analyzing and recognizing the user's emotions by applying an emotion recognition algorithm, means for generating optimal suggestions and responses based on the analysis results of the generative AI model and the emotion data, and means for providing the generated suggestions and responses to the user audio or visually, thereby enabling accurate analysis of the user's input and providing personalized responses and suggestions that take emotions into consideration.

[1655] "Voice or textual input means" refers to an interface for receiving voice or textual input from a user and processing that data.

[1656] "Speech recognition software" refers to a program used to convert voice data into text data. A specific example is automatic speech recognition (ASR) technology.

[1657] A "server" refers to a computer system connected to a network for processing requests from multiple clients.

[1658] A "generative AI model" refers to an artificial intelligence algorithm that analyzes user input data and generates appropriate responses or suggestions based on that data. A specific example is a natural language processing model.

[1659] "Means of capturing using a camera and microphone" refers to a method of using a camera and microphone to capture a user's facial expressions and tone of voice.

[1660] "Emotion recognition algorithm" refers to an algorithm that analyzes acquired data (facial expressions and voice tone) to recognize the user's emotional state (e.g., joy, sadness, tension, etc.).

[1661] "Emotion engine" refers to a software or hardware configuration for executing emotion recognition algorithms.

[1662] "Means for generating suggestions or responses" refers to a method for creating appropriate suggestions or responses for a user based on the analysis results and emotional data.

[1663] "Audio or visual means" refers to a method for communicating generated suggestions or responses to the user, such as by audio output or display.

[1664] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. The system receives user input in the form of voice or text, analyzes the input, recognizes the user's emotion, and then provides appropriate responses and suggestions.

[1665] System configuration

[1666] 1. Accepting user input

[1667] Users can enter questions or requests in voice or text format, for example, "What should I wear today?"

[1668] The terminal (robot) receives this voice and converts it into text data using voice recognition software (for example, a voice recognition API). If the input is in text format, it is accepted by the system as is.

[1669] 2. Data submission and analysis

[1670] The terminal transmits the converted text data to the server.

[1671] The server inputs the received text data into a generative AI model (specifically, a natural language processing model) and analyzes the question. For example, it identifies the theme "wearing clothes."

[1672] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data APIs, fashion database APIs). For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1673] 3. Emotion recognition

[1674] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1675] The device sends the acquired data to an emotion engine, which applies emotion recognition algorithms (e.g., machine learning algorithms) to determine, for example, whether the user is tense or relaxed.

[1676] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates information such as "the user is nervous."

[1677] 4. Generating the Response

[1678] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1679] 5. User Feedback

[1680] The terminal (robot) provides the generated suggestions and responses to the user both audibly and visually, for example, saying "How about this casual jacket?" and showing an image on the display.

[1681] Specific examples

[1682] Coordination suggestions at apparel shops

[1683] The user asks aloud, "I'm going to dinner with friends today. What kind of clothes should I wear?"

[1684] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[1685] The server inputs the text data into a generative AI model (natural language processing model) to analyze the user's intentions, and then generates optimal outfits based on weather information and the latest fashion trends (fashion database API).

[1686] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional information such as whether the user is nervous or relaxed.

[1687] The emotion engine analyzes the emotion information to generate emotion data, which is then sent to the server.

[1688] Based on the acquired emotional data, the server adjusts its suggestions to match the user's preferences, asking aloud, "How about a relaxed, casual outfit?" while displaying specific outfit images on the screen.

[1689] The user can review the suggestions and ask further questions or make requests, for example, "Do you have a more casual jacket?"

[1690] Menu suggestions at restaurants

[1691] The user asks verbally, "What are your recommendations for today?"

[1692] The terminal (robot) converts this voice into text data using a voice recognition API and sends it to the server.

[1693] The server inputs the text data into a generative AI model (natural language processing model) and analyzes the question. It then generates recommended menu items by referencing menu information and seasonal special dishes (special dish database).

[1694] The device analyzes the user's facial expressions and voice tone using an emotion recognition algorithm (machine learning algorithm) to obtain emotional data.

[1695] The emotion engine adjusts the content of the suggestions based on the acquired emotion data. It suggests aloud, "Today is a special day, so how about this special menu?" while showing visual images.

[1696] As described above, the present invention is a system that can realize natural dialogue with a user and provide appropriate and personalized suggestions and responses.

[1697] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1698] Step 1: Accept user input

[1699] The user types in a question or request through voice or text, for example, "What should I wear today?"

[1700] Input: User voice or text data

[1701] The terminal (robot) receives voice using a microphone and converts the voice into text data using voice recognition software (e.g., voice recognition API). If the input is text, it is accepted by the system as is.

[1702] Output: Converted text data

[1703] Specific behavior:

[1704] The device collects the user's voice through a microphone.

[1705] Call the speech recognition API and convert the speech data into text data.

[1706] The converted text data is temporarily stored.

[1707] Step 2: Sending data

[1708] The device then sends the converted text data to the server.

[1709] Input: Text data stored on the device

[1710] Output: Text data sent to the server

[1711] Specific behavior:

[1712] The converted text data is sent to the server via the network.

[1713] Verify that the submission was successful.

[1714] Step 3: Analyze the data

[1715] The server inputs the received text data into a generative AI model (e.g., a natural language processing model) and analyzes the question content.

[1716] Input: Text data

[1717] Output: Parsed question topic (e.g. "Outfit suggestions")

[1718] Specific behavior:

[1719] The server preprocesses the received text data and generates a prompt (e.g., "The user is asking, 'What should I wear today?' Please generate an appropriate response.").

[1720] The prompt sentence is input into the generative AI model, and the model begins analysis.

[1721] As a result of the analysis, the topic of the question and related information are identified.

[1722] Step 4: Gather information

[1723] The generative AI model collects and analyzes information from relevant databases and external APIs (e.g., weather data API, fashion database API).

[1724] Input: Question topic, related information retrieved from databases and external APIs

[1725] Output: Analysis results (e.g., optimal outfits based on the latest fashion trends or weather data)

[1726] Specific behavior:

[1727] The generative AI model calls database APIs or external APIs to obtain the necessary information.

[1728] Based on the acquired information, the model performs analysis and generates coordination suggestions.

[1729] Step 5: Emotion Recognition

[1730] The device uses a camera and microphone to capture the user's facial expressions and voice tone.

[1731] Input: Camera video data, audio tone data

[1732] Output: Captured facial expression data and voice tone data

[1733] Specific behavior:

[1734] The device captures the user's facial expressions with a camera and collects voice tones with a microphone.

[1735] The collected facial expression data and voice tone data are temporarily stored.

[1736] Step 6: Analyze the sentiment data

[1737] The device sends the acquired data to the emotion engine and applies an emotion recognition algorithm (e.g., a machine learning algorithm).

[1738] Input: facial expression data, voice tone data

[1739] Output: Emotion data (e.g., "The user is nervous")

[1740] Specific behavior:

[1741] Facial expression data and voice tone data are sent to the emotion engine.

[1742] An emotion recognition algorithm is used to analyze emotions and generate emotion data.

[1743] The generated emotion data is sent to the server.

[1744] Step 7: Generate a response

[1745] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user.

[1746] Input: Analysis results (e.g., outfit suggestions), emotion data

[1747] Output: Optimal suggestion or response (e.g., "Suggestions for relaxing casual clothing")

[1748] Specific behavior:

[1749] Integrate analysis results with emotion data.

[1750] It takes into account the user's emotional state to instruct the generative AI model on how to generate a response.

[1751] Generate optimal proposals and responses and send them to the device.

[1752] Step 8: User feedback

[1753] The terminal (robot) provides the generated suggestions and responses to the user audibly and visually.

[1754] Input: Best suggestion or response

[1755] Output: Audio and visual suggestions and responses provided to the user

[1756] Specific behavior:

[1757] The terminal uses speech synthesis software to audibly output the generated suggestions.

[1758] At the same time, an image of the coordinated outfit is displayed on the display.

[1759] The user can review the information provided and ask further questions or requests, for example, "Do you have a more casual jacket?"

[1760] (Application example 2)

[1761] 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."

[1762] In today's brick-and-mortar stores, it is not easy for users to receive accurate and personalized suggestions when selecting products or using services. Conventional systems have difficulty not only analyzing user input, but also properly recognizing the user's emotional state and making optimal suggestions based on this. This has resulted in the problem of insufficient effective customer service and service provision.

[1763] 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 accepting user input in voice or text format; means including a generative AI model for analyzing the user input; means for recognizing the user's emotion by applying an emotion recognition algorithm; means for generating optimal suggestions and responses based on the user's input and emotion data; means for providing the generated suggestions and responses to the user audio or visually; and means for analyzing the user's facial expressions and tone of voice using a camera and microphone when the user asks a question in a physical store, and correcting the generative AI model based on the emotion data to make optimal suggestions. This enables personalized suggestions and responses based on the user's emotional state.

[1764] "User Input" means a question or request provided by a User in voice or text form.

[1765] A "generative AI model" is an artificial intelligence model that analyzes user input, understands text, voice, images, and video, and generates optimal suggestions and responses.

[1766] The "emotion recognition algorithm" is an algorithm that uses a camera and microphone to analyze a user's facial expressions and tone of voice to understand their emotional state.

[1767] "Emotional data" is data that indicates a user's emotional state as analyzed by an emotion recognition algorithm.

[1768] "Suggestions and responses" refer to optimal information and suggestions derived by the generative AI model based on user input and emotional data.

[1769] "Means for providing audio or visual information to the user" refers to means for conveying generated suggestions or responses to the user through audio output or visual information such as a display.

[1770] A "brick and mortar store" is a physical location where consumers can visit and consume products or services.

[1771] "Camera and microphone" refers to video and audio input devices for capturing the user's facial expressions and voice tone.

[1772] The present invention is a system for analyzing user input and generating optimal suggestions and responses, which is realized by combining a generative AI model with an emotion recognition algorithm and an emotion engine. Specifically, this system is implemented as follows.

[1773] System configuration

[1774] 1. Accepting user input

[1775] The device provides an interface through which the user can ask questions or make requests in the form of voice or text, for example, by saying, "What should I wear today?"

[1776] The device accepts voice input and converts it into text data, or accepts text input as is.

[1777] 2. Data submission and analysis

[1778] The terminal transmits the converted text data to the server.

[1779] The server inputs the received text data into a generative AI model and analyzes the question content, for example, identifying the theme "wearing clothes."

[1780] The generative AI model understands text, voice, images, and video, and collects and analyzes information from relevant databases and external APIs. For example, it generates optimal outfits based on the latest fashion trends and weather data.

[1781] 3. Emotion recognition

[1782] The device uses a camera and microphone to capture the user's facial expressions and tone of voice, which allows it to understand the user's emotions.

[1783] The device sends data captured by the camera and microphone to an emotion engine that applies emotion recognition algorithms to determine, for example, whether the user is tense or relaxed.

[1784] The emotion engine generates emotion data as a result of the analysis and sends it to the server. For example, it generates data indicating that the user is nervous.

[1785] 4. Generating the Response

[1786] The server combines the analysis results of the generative AI model with emotional data to generate optimal suggestions and responses for the user, such as suggesting "casual clothing that is comfortable to wear."

[1787] 5. User Feedback

[1788] The device will then provide the generated suggestions and responses to the user both audibly and visually, for example, asking a question aloud, "What do you think of this casual jacket?", while showing an image on the display.

[1789] Hardware and software used

[1790] 1. Hardware

[1791] camera

[1792] microphone

[1793] Smart glasses, head-mounted displays, tablets

[1794] 2. Software

[1795] Emotion recognition libraries (e.g., Affectiva)

[1796] Generative AI models (e.g., OpenAI GPT-4)

[1797] Speech processing engine (e.g. Google Speech-to-Text)

[1798] GUI library for display (e.g. Qt)

[1799] Examples of concrete examples and prompts

[1800] Examples:

[1801] When a user asks "What's your recommended outfit for today?" and the emotion recognition system determines that the answer is "Relaxed," the prompt is:

[1802] User Question: What's your go-to outfit for today?

[1803] The user is relaxed.

[1804] Please make appropriate suggestions.

[1805] Thus, the present invention is a system that can realize natural dialogue with the user and provide appropriate and personalized suggestions and responses.

[1806] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1807] Step 1:

[1808] A user inputs a question or request to a terminal in the form of voice or text. This is done through a voice interface or a text input interface. For example, the question "What should I wear today?" is input.

[1809] Input: User voice or text question

[1810] Output: Audio data or text data

[1811] Step 2:

[1812] The device converts the voice input into text data. If the input is voice, a speech processing engine (e.g., Google Speech-to-Text) converts the voice into text. If the input is text, this step is skipped.

[1813] Input: Audio data

[1814] Output: Text data

[1815] Step 3:

[1816] The device sends text data to the server, which receives it and inputs it into the generative AI model.

[1817] Input: Text data

[1818] Output: Request for analysis results

[1819] Step 4:

[1820] The server analyzes the text data and understands the question using a generative AI model (e.g., OpenAI GPT-4). The generative AI model collects and analyzes relevant information to generate optimal suggestions, taking into account, for example, the latest fashion trends and weather data.

[1821] Input: Text data

[1822] Output: Initial proposal data

[1823] Step 5:

[1824] The device uses a camera and microphone to capture the user's facial expressions and voice tone, and this data is sent to an emotion recognition algorithm (e.g., Affectiva) to analyze the user's emotional state.

[1825] Input: User's face image and voice tone

[1826] Output: Emotion data

[1827] Step 6:

[1828] The emotion engine generates emotion data analyzed by an emotion recognition algorithm and sends it to the server, which then provides emotion information such as "the user is nervous."

[1829] Input: Facial image data and voice tone data

[1830] Output: Parsed emotion data

[1831] Step 7:

[1832] The server combines the analysis results of the generated AI model with emotional data to generate optimal suggestions and responses. For example, it may suggest "casual clothing that is comfortable to wear."

[1833] Input: Initial proposal data and emotion data

[1834] Output: Optimized proposal data

[1835] Step 8:

[1836] The device provides the generated suggestions and responses to the user both audibly and visually. Using a voice output engine and a display, the device displays an image on the screen while providing a voice prompt such as, "How about this casual jacket?"

[1837] Input: Optimized proposal data

[1838] Output: Audio audio output and visual display

[1839] In this way, each step processes specific input data and produces an appropriate output based on that data, allowing for personalized suggestions and responses to be provided to the user.

[1840] 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.

[1841] 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.

[1842] 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.

[1843] 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.

[1844] 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.

[1845] 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.

[1846] 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).

[1847] 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.

[1848] 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."

[1849] 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.

[1850] 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).

[1851] 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.

[1852] 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.

[1853] 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.

[1854] 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.

[1855] 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.

[1856] 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.

[1857] 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.

[1858] 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.

[1859] 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.

[1860] 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.

[1861] The following is further disclosed regarding the above embodiment.

[1862] (Claim 1)

[1863] a means for accepting user input in the form of speech or text;

[1864] a means for interacting with a server containing a generative AI model for analyzing user input;

[1865] means for applying an emotion recognition algorithm to recognize the emotion of the user;

[1866] a means for generating optimal suggestions and responses based on user input and sentiment data;

[1867] a means of providing generated suggestions and responses to the user audibly or visually; and

[1868] A system including:

[1869] (Claim 2)

[1870] 10. The system of claim 1, wherein the generative AI model includes the ability to understand and analyze text, speech, images, and video.

[1871] (Claim 3)

[1872] 10. The system of claim 1, wherein the emotion recognition algorithm comprises means for analyzing a user's facial expression and / or vocal tone to generate emotion data.

[1873] "Example 1"

[1874] (Claim 1)

[1875] a means for accepting user input in the form of speech or text;

[1876] means for converting voice input into text data;

[1877] a means for interacting with a server containing a generative AI model for analyzing user input;

[1878] means for applying an emotion recognition algorithm to analyze a user's facial expression and tone of voice to generate emotion data;

[1879] a means for generating optimal suggestions and responses based on user input and sentiment data;

[1880] a means of providing generated suggestions and responses to the user audibly or visually; and

[1881] A system including:

[1882] (Claim 2)

[1883] 10. The system of claim 1, wherein the generative AI model includes the ability to understand and analyze text, speech, images, and video.

[1884] (Claim 3)

[1885] 10. The system of claim 1, wherein the emotion recognition algorithm comprises means for analyzing a user's facial expression and / or vocal tone to generate emotion data.

[1886] "Application Example 1"

[1887] (Claim 1)

[1888] a means for accepting user input in the form of speech or text;

[1889] a means for interacting with a server containing a generative AI model for analyzing user input;

[1890] means for applying an emotion recognition algorithm to recognize the emotion of the user;

[1891] a means for generating optimal suggestions and responses based on user input and sentiment data;

[1892] a means of providing generated suggestions and responses to the user audibly or visually; and

[1893] A means for correcting the suggestions based on the user's emotions;

[1894] A system including:

[1895] (Claim 2)

[1896] 10. The system of claim 1, wherein the generative AI model includes the ability to understand and analyze text, speech, images, and video.

[1897] (Claim 3)

[1898] 10. The system of claim 1, wherein the emotion recognition algorithm comprises means for analyzing a user's facial expression and / or vocal tone to generate emotion data.

[1899] "Example 2: Combining Emotion Engines"

[1900] (Claim 1)

[1901] a means for accepting user input in the form of speech or text;

[1902] using speech recognition software to convert received speech data into text data;

[1903] means for transmitting the converted text data to a server;

[1904] a means for analyzing user input in conjunction with a server containing the generative AI model;

[1905] A means for capturing a user's facial expressions and tone of voice using a camera and microphone;

[1906] means for applying an emotion recognition algorithm to analyze and recognize the user's emotions;

[1907] A means for generating optimal suggestions and responses based on the analysis results of the generative AI model and emotional data; and

[1908] a means of providing generated suggestions and responses to the user audibly or visually; and

[1909] A system including:

[1910] (Claim 2)

[1911] 10. The system of claim 1, wherein the generative AI model includes the ability to understand text, speech, images, and video, and to collect and analyze information from external databases and APIs.

[1912] (Claim 3)

[1913] 10. The system of claim 1, wherein the emotion recognition algorithm comprises means for analyzing a user's facial expression and tone of voice and generating emotion data from the acquired data.

[1914] "Application example 2 when combining emotion engines"

[1915] (Claim 1)

[1916] a means for accepting user input in the form of speech or text;

[1917] a means for interacting with a server containing a generative AI model for analyzing user input;

[1918] means for applying an emotion recognition algorithm to recognize the emotion of the user;

[1919] a means for generating optimal suggestions and responses based on user input and sentiment data;

[1920] a means of providing generated suggestions and responses to the user audibly or visually; and

[1921] When a user asks a question in a physical store, a camera and microphone are used to analyze the user's facial expressions and tone of voice, and the generated AI model is adjusted based on the emotional data to make optimal suggestions.

[1922] A system including:

[1923] (Claim 2)

[1924] 10. The system of claim 1, wherein the generative AI model includes the ability to understand and analyze text, speech, images, and video.

[1925] (Claim 3)

[1926] 10. The system of claim 1, wherein the emotion recognition algorithm comprises means for analyzing a user's facial expression and / or vocal tone to generate emotion data. [Explanation of symbols]

[1927] 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 accepting user input in the form of speech or text; a means for interacting with a server containing a generative AI model for analyzing user input; means for applying an emotion recognition algorithm to recognize the emotion of the user; a means for generating optimal suggestions and responses based on user input and sentiment data; a means of providing generated suggestions and responses to the user audibly or visually; and A system including:

2. The system of claim 1 , wherein the generative AI model includes the ability to understand and analyze text, speech, images, and video.

3. 2. The system of claim 1, wherein the emotion recognition algorithm includes means for analyzing a user's facial expression and tone of voice to generate emotion data.

Images