System

Abstract

A system is provided.SOLUTION: A system comprising: a music provider comprising a generative artificial intelligence model for generating a music playlist based on a user's preferences; an interactor for interacting with the user to update the music playlist in real-time; a profile collector for obtaining the user's current mood and activity; and an input for inputting data collected by the profile collector into the generative artificial intelligence model.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 music streaming services have the drawback of being time-consuming and inefficient for users to find music that matches their preferences. They also face the challenge of making it difficult for users to smoothly acquire music that matches their mood and activity status. Furthermore, direct interaction with a DJ is typically only available at live events, limiting the ability to provide a personalized music experience. The present invention aims to solve these issues and provide users with a music experience that is optimized for them. [Means for solving the problem]

[0005] The present invention provides a system including a music providing means having a generative artificial intelligence model that generates a music playlist based on a user's preferences, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, and an input means that inputs data collected by the profile collecting means into the generative artificial intelligence model. The generative artificial intelligence model also includes an algorithm that analyzes the user's past music playback history and recommends new music based on the user's preferences. Furthermore, the means for interacting with the generative artificial intelligence model has a function for accepting requests from the user via text and voice input, thereby providing the user with a personalized music experience.

[0006] A "generative artificial intelligence model" refers to an artificial intelligence technology that generates appropriate music playlists based on user preferences and past playback history.

[0007] "Music Provider" refers to technologies and systems for providing users with music playlists generated by a generative artificial intelligence model.

[0008] "Interaction Means" refers to technologies and systems for interacting with users in real time and updating music playlists based on user requests and feedback.

[0009] "Profile collection means" refers to technologies and systems for collecting information such as a user's current mood, activity status, and past music playback history.

[0010] "Input Means" refers to the technology and systems for inputting data collected by the Profile Collection Means into the generative artificial intelligence model.

[0011] "Music streaming service" refers to a service that distributes and plays music in real time over the Internet.

[0012] "User preferences" refers to the user's personal preferences, such as the music genres, artists, and songs they prefer.

[0013] Unlike a live event, a "private DJ experience" refers to providing a music playlist optimized for each user and customizing the music experience individually through interaction with the user. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] Collecting user profiles

[0036] 1. A user starts the music streaming application and begins using the system by logging in.

[0037] Example: A user opens an app and enters their login details to access their account.

[0038] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[0039] Example: An app pops up a question to the user, such as "How are you feeling right now?" or "What are your current activities?"

[0040] 3. The user enters the information into the terminal.

[0041] Example: A user enters information such as "I want to relax" or "I'm exercising."

[0042] 4. The device retrieves past music playback history from local storage and sends it to the server.

[0043] Example: An app gets a list of previously played songs and sends it to a server.

[0044] Data analysis and playlist generation

[0045] 1. The server receives the user's preferences and current state data.

[0046] Example: Receiving data such as "User ID: 12345, current mood: relaxed, activity: exercising."

[0047] 2. The server loads the generative artificial intelligence model and inputs the user profile.

[0048] Example: Providing profile information such as "relaxed" or "exercising" to an AI model.

[0049] 3. The server uses this data to generate a music playlist that best suits the user's current situation.

[0050] Example: Generate a playlist of relaxing and exercise-friendly music such as "Lo-fi Hip Hop" and "House Music."

[0051] 4. The server sends the generated playlist to the user's device.

[0052] Example: The generated playlist is sent to the app and displayed to the user.

[0053] Real-time playlist updates

[0054] 1. Listen to songs from a user-provided playlist.

[0055] Example: A user starts playing a song from a recommended playlist.

[0056] 2. The user inputs feedback about the song being played into the device, such as whether they like it or not, or if they want to skip it.

[0057] Example: Enter feedback such as "I like this song" or "I want to hear the next song."

[0058] 3. The device sends the feedback to the server.

[0059] Example: An app sends feedback information to a server in real time.

[0060] 4. The server regenerates or updates the playlist based on the feedback.

[0061] Example: Adding or changing songs in a playlist based on new feedback.

[0062] 5. The server sends the updated playlist back to the user's device.

[0063] Example: The updated playlist is reflected and displayed on the user's device.

[0064] User interaction

[0065] 1. The user makes a request or asks the AI ​​DJ a question, such as "What song do you recommend next?"

[0066] Example: A user types the question "What song should I play next?"

[0067] 2. The device sends the user's request or question to the server.

[0068] Example: An app sends a message from the user to a server.

[0069] 3. The server uses the AI ​​DJ model to generate an appropriate response.

[0070] Example: An AI DJ generates messages such as, "The song I recommend for your next relaxation session is XX."

[0071] 4. The server generates a response and sends it to the user's device.

[0072] Example: A response message is displayed in the user's app.

[0073] 5. The device displays or plays the AI ​​DJ's response to the user.

[0074] For example: The generated response is displayed or played in text or audio format.

[0075] This system allows users to get the best possible music experience based on their preferences and current situation, and allows for a more personalized music experience through real-time feedback and interaction.

[0076] The processing flow will be explained below.

[0077] Program processing steps

[0078] ---

[0079] Collecting user profiles

[0080] Step 1:

[0081] A user launches a music streaming application and logs in.

[0082] Specific behavior: Enter your username and password on the app's login screen and submit your authentication information.

[0083] Step 2:

[0084] The device asks the user about their current mood and activity.

[0085] Specific behavior: The app prompts the user with questions such as "How are you feeling right now?" and "What are you currently doing?"

[0086] Step 3:

[0087] The user inputs their current mood and activity status into the terminal.

[0088] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[0089] Step 4:

[0090] The device retrieves past music playback history from local storage and sends it to the server.

[0091] Specific operation: Reads playback history data from local storage and sends it to the server along with collected mood and activity information.

[0092] ---

[0093] Data analysis and playlist generation

[0094] Step 5:

[0095] The server receives the user's current mood, activity, and playback history data.

[0096] Specific operation: The server saves the received data in the database and begins analysis.

[0097] Step 6:

[0098] The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[0099] Specific operation: Create an instance of the AI ​​model and pass profile data such as "relaxed" or "exercising" as input parameters.

[0100] Step 7:

[0101] The server uses an AI model to generate a playlist based on the user's profile.

[0102] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile.

[0103] Step 8:

[0104] The server transmits the generated playlist to the user's terminal.

[0105] Specific operation: The selected song list is structured in JSON format and sent to the user's device.

[0106] ---

[0107] Real-time playlist updates

[0108] Step 9:

[0109] The terminal displays the generated playlist to the user.

[0110] Specific behavior: Parse the transmitted playlist data and display it in the app interface.

[0111] Step 10:

[0112] The user plays a song from the playlist.

[0113] Specific behavior: The song selected by the user will be played within the app.

[0114] Step 11:

[0115] The user inputs feedback about the song into the terminal.

[0116] Specific action: Enter feedback such as "I like this song" or "I want to hear the next song" into the input form.

[0117] Step 12:

[0118] The terminal transmits the feedback from the user to the server.

[0119] Specific behavior: The input feedback is sent to the server in real time.

[0120] Step 13:

[0121] The server regenerates or updates the playlist based on the feedback.

[0122] Specific behavior: Analyzes the received feedback data and performs processing to update the playlist.

[0123] Step 14:

[0124] The server sends the updated playlist to the user's terminal.

[0125] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[0126] ---

[0127] User interaction

[0128] Step 15:

[0129] Users make requests and ask questions to the AI ​​DJ.

[0130] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[0131] Step 16:

[0132] The terminal transmits the user's requests and questions to the server.

[0133] Specific operation: Sends the entered message data to the server.

[0134] Step 17:

[0135] The server uses an AI DJ model to generate appropriate responses.

[0136] Specific operation: The AI ​​DJ model analyzes the request content and generates an appropriate response message.

[0137] Step 18:

[0138] The server generates a response and sends it to the user's terminal.

[0139] Specific operation: The generated response message is sent to the user's terminal.

[0140] Step 19:

[0141] The terminal displays or plays the AI ​​DJ's responses to the user.

[0142] Specific behavior: The response message sent is displayed on the screen or played aloud.

[0143] Example 1

[0144] 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."

[0145] Conventional music streaming systems have struggled to provide real-time music playlists that respond to users' preferences, current moods, and activity status. Furthermore, they lack the ability to respond immediately to user feedback and requests, making it impossible to provide an optimized music experience for each individual user. The present invention aims to solve these problems and provide users with a more personalized music experience.

[0146] 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.

[0147] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model. This makes it possible to generate and update an optimal music playlist in real time based on the user's preferences and current status and provide it to the user.

[0148] "User" refers to an individual who uses the music streaming system.

[0149] "Preferences" refers to a user's musical tastes and tendencies.

[0150] "Mood" refers to the emotion or mental state a user is feeling at a particular time.

[0151] An "activity" is an action or task that a user is performing at a particular time.

[0152] "Music Playlist" means a list of songs organized for sequential playback by a user.

[0153] A "generative artificial intelligence model" refers to an artificial intelligence technology that generates new information or recommendations based on input data.

[0154] "Music provision means" refers to a mechanism that uses a generative artificial intelligence model to provide users with music playlists.

[0155] "Interaction means" refers to a mechanism by which users and systems communicate with each other.

[0156] "Profile Collection Measures" refers to mechanisms that collect information about a user's preferences, moods, and activities.

[0157] "Input means" refers to a mechanism for inputting data collected by the profile collection means into the generative artificial intelligence model.

[0158] "Server" refers to the computer system that processes user data and runs the generative artificial intelligence model.

[0159] This invention relates to a system that generates and updates an optimal music playlist in real time according to a user's preferences, current mood, and activity status. Specific embodiments of this system are described in detail below.

[0160] The system is mainly composed of three entities: a server, a terminal, and a user.

[0161] Collecting user profiles

[0162] A user begins using the system by launching a music streaming application and logging in. The user opens the app on a device such as a smartphone or tablet and enters their login information for authentication. If authentication is successful, the device displays a pop-up window asking about the user's current mood and activity. When the user enters information such as "I want to relax" or "I'm exercising," the device collects this information and retrieves past music playback history from local storage.

[0163] Data analysis and playlist generation

[0164] The data collected by the device is sent to the server. The server loads a generative artificial intelligence model and inputs the user profile based on the received user preferences and current state data. The generative AI model uses this data to generate a music playlist that is optimal for the user's current situation. The generated playlist is then sent back to the user's device, where it is displayed.

[0165] Real-time playlist updates

[0166] While listening to songs in a playlist, a user can input feedback about the song being played into the device. If the user inputs feedback such as "I like this song" or "I want to hear the next song," the device sends this information to the server in real time. The server regenerates or updates the playlist based on this feedback and sends the new playlist to the user's device.

[0167] User interaction

[0168] Users can make requests or questions to the AI ​​DJ. For example, if they input a question like "What's the next song you recommend?", the device will send this request to the server. The server will use the AI ​​DJ model to generate an appropriate response and send it to the user's device. The device will then display or play this response in text or audio format.

[0169] The system allows users to get the best possible music experience in real time based on their preferences and current situation, and further personalizes the music experience through user feedback and interaction.

[0170] Examples of prompt statements

[0171] Prompt: "What are the steps for a system that allows a user to enter their current mood and activity in a music streaming app and have AI generate the perfect playlist?"

[0172] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0173] Step 1:

[0174] The user launches the app and logs in

[0175] A user launches a music streaming app on a device such as a smartphone or tablet and enters their login information (email address, password). The device sends this input data to the server, which then authenticates the user. If authentication is successful, the server returns an authentication success message to the device, allowing the user to access their account.

[0176] Input: Email address, password

[0177] Output: Authentication success message

[0178] Step 2:

[0179] The device asks about the user's mood and activity

[0180] After logging in, the device displays a pop-up window asking the user questions such as "How are you feeling now?" and "What is your current activity?" To make it easier for the user to answer, options (e.g., relaxing, exercising, etc.) are displayed. These options may also be provided by the server.

[0181] Input: Authentication success message

[0182] Output: Mood and activity question popup

[0183] Step 3:

[0184] The user enters information

[0185] The user selects and inputs their mood and activity (e.g., "I want to relax" or "I'm exercising") from a pop-up window on the device, and the device immediately sends this information to the server.

[0186] Input: Questions about mood and activity

[0187] Output: User's mood and activity information

[0188] Step 4:

[0189] The device acquires past music playback history and sends it to the server.

[0190] The device searches for and retrieves past music playback history from its internal local storage, and the retrieved data is sent to the server.

[0191] Input: User's mood and activity information

[0192] Output: Past music playback history

[0193] Step 5:

[0194] The server receives the user's preferences and current state data.

[0195] The server receives data packets from the device about the user's mood, activity, and playback history, stores them in a database, and prepares them for analysis.

[0196] Input: User's mood and activity information, past music playback history

[0197] Output: Data stored in the database

[0198] Step 6:

[0199] The server loads the generative artificial intelligence model and inputs the user profile.

[0200] The server loads the generative AI model file from disk and expands it into memory. Data on the user's current mood, activity, and playback history is input into the generative AI model, and analysis begins.

[0201] Input: Data stored in a database

[0202] Output: User profile fed into the AI ​​model

[0203] Step 7:

[0204] The server generates a music playlist that best suits the user's current situation.

[0205] The server uses an AI model to select the best songs based on the user's input data and generate a playlist, which is then packaged into packets and sent to the device.

[0206] Input: User profile fed into the AI ​​model

[0207] Output: Generated music playlist

[0208] Step 8:

[0209] The server sends the generated playlist to the device.

[0210] The server assembles the generated playlist into a data packet and sends it to the user's terminal, which receives it and displays it to the user.

[0211] Input: Generated music playlist

[0212] Output: Playlist displayed on device

[0213] Step 9:

[0214] Listen to songs from a user-provided playlist

[0215] The user presses the play button on the playlist screen within the app to begin listening to the song, and the device responds by streaming the song data from the server.

[0216] Input: Playlist displayed on device

[0217] Output: The song being played

[0218] Step 10:

[0219] Users provide feedback on songs

[0220] Users can input feedback by tapping buttons on the playback screen such as "I like this song" or "I want to listen to the next song." The device then sends this feedback information to the server in real time.

[0221] Input: User feedback

[0222] Output: Feedback information sent to the server

[0223] Step 11:

[0224] The server regenerates or updates the playlist based on the feedback

[0225] The server re-runs the AI ​​model based on the received feedback, updates the playlist, and reconstructs the newly generated playlist and sends it to the device.

[0226] Input: Feedback information sent to the server

[0227] Output: Regenerated music playlist

[0228] Step 12:

[0229] The server sends the updated playlist to the device.

[0230] The server packages the updated playlist into a data packet and sends it to the user's terminal, which receives it and displays the new playlist.

[0231] Input: Regenerated music playlist

[0232] Output: Updated playlist displayed on device

[0233] Step 13:

[0234] Users make requests and ask questions to the AI ​​DJ

[0235] Users can type or speak using the in-app chat box or voice assistant function, asking, "What's your recommendation for the next song?"

[0236] Input: User requests or questions

[0237] Output: Requests and questions typed into the terminal

[0238] Step 14:

[0239] The device sends requests and questions to the server

[0240] The device sends the user's request or question as a data packet to the server, which receives the data and begins analyzing it.

[0241] Input: Requests or questions typed into the device

[0242] Output: The request or question sent to the server

[0243] Step 15:

[0244] The server generates a response using the AI ​​DJ model

[0245] The server runs the AI ​​DJ model to generate appropriate responses to requests and questions, which are then packaged into data packets and sent to the device.

[0246] Input: The request or question sent to the server

[0247] Output: The generated response message

[0248] Step 16:

[0249] Sends the server-generated response to the terminal

[0250] The server generates a response message and sends it as a data packet to the terminal, which receives it and displays or plays it audibly to the user.

[0251] Input: The generated response message

[0252] Output: The greeting message displayed or played on the terminal

[0253] (Application example 1)

[0254] 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."

[0255] Conventional food delivery systems have struggled to recommend the best food and restaurant in real time based on a user's preferences and current mood. Furthermore, they lacked the technology to instantly update recommendations based on user feedback. This resulted in a lack of consistency and individualization in the user experience.

[0256] 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.

[0257] In this invention, the server includes a recommendation means equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences, an interaction means that updates the recommendations in real time by interacting with the user, a profile collection means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collection means into the generative artificial intelligence model. This makes it possible to recommend optimal dishes and restaurants in real time based on the user's mood and activity status, and to instantly update the recommendations in response to feedback.

[0258] A "recommendation vehicle" is a vehicle equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences.

[0259] An "interaction means" is a means for updating recommendations in real time by interacting with the user.

[0260] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[0261] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[0262] A "generative artificial intelligence model" is an artificial intelligence model that generates optimal recommendations based on user preferences.

[0263] "Past order history" refers to the history of orders previously placed by the user.

[0264] An "algorithm" is a set of steps or rules for solving a specific problem by following a set of steps.

[0265] The system for implementing this invention mainly comprises a user terminal, a server, and a generative artificial intelligence model. The user terminal includes a mobile device such as a smartphone, and the server is located in a remote data center. The system has the function of collecting user profile information and providing optimal recommendations based on that information.

[0266] First, when a user logs in to the application using their smartphone, the device asks about the user's current mood and activity status. When the user enters information such as "I want to relax" or "I'm at a party," the device retrieves this information from local storage and sends it to the server.

[0267] The server receives the user's current status and past order history. Specifically, data such as "User ID: 12345, current mood: relaxed, activity: partying" is received and input into a generative AI model. This generative AI model generates optimal recommendations based on the user profile.

[0268] The generative AI model then uses the user's profile information to generate a list of recommended dishes and restaurants. For example, if a user wants to relax, snacks and desserts might be recommended, while if they're at a party, pizza and alcoholic drinks might be recommended.

[0269] Once the recommendation list is generated, the server sends the information to the user's device and displays it on the device. The user selects from the recommendation list and enters feedback into the device. For example, the user may enter feedback such as "I like this dish" or "I'd like to try a different dish next time." The device then sends this feedback to the server in real time, and the server again uses the generative artificial intelligence model to update the recommendation list.

[0270] The hardware used includes smartphones (user devices) and servers (remote data centers), and the software includes Python, Flask, generative artificial intelligence model libraries (e.g., some_ai_lib), JSON, HTTP, etc.

[0271] For example, if the prompt information "Current opinion: Relax, Activity: Partying" is entered, the generative artificial intelligence model will provide a recommendation based on this prompt information, such as "The next recommended dish is sushi and a tapioca drink."

[0272] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0273] Step 1:

[0274] A user logs in to an application on a smartphone. The user enters login information and accesses the application. The device receives this login information and performs user authentication. The input is the login information, and the output is the authentication status (success or failure).

[0275] Step 2:

[0276] If authentication is successful, the device displays a pop-up asking the user about their current mood and activity status. The user inputs information such as "I want to relax" or "I'm at a party." The input is the user's mood and activity status, and the output is a data format containing this information.

[0277] Step 3:

[0278] The device retrieves the user's current mood and activity status from local storage and sends it to the server. The input is the data in local storage and newly entered information, and the output is the data format (e.g., JSON format) sent to the server.

[0279] Step 4:

[0280] The server receives the user's current mood, activity status, and past order history and inputs them into a generative artificial intelligence model. The inputs are the user's current mood, activity status, and past order history, and the output is the profile data that is fed into the model.

[0281] Step 5:

[0282] A generative artificial intelligence model uses a user's profile information to generate a list of recommendations for the most suitable dishes and restaurants. The input is the user profile information, and the output is a list of recommendations. Specifically, the AI ​​model processes the data by analyzing past history and current status, and then uses a matching algorithm to identify recommended items.

[0283] Step 6:

[0284] The server sends the generated recommendation list to the user terminal. The input is the generated recommendation list, and the output is the recommendation list displayed on the user terminal. The user terminal displays this list on its screen.

[0285] Step 7:

[0286] The user selects from the recommendation list and inputs feedback into the terminal, for example, "I like this dish" or "I'd like to try a different dish next time." The input is the user's feedback, and the output is the feedback data sent to the server.

[0287] Step 8:

[0288] The device sends real-time feedback to the server, which then uses the generative AI model to update the recommendation list. The input is the feedback data, and the output is an updated recommendation list. Specifically, the AI ​​model re-runs optimization based on the new feedback to calculate the optimal recommendations.

[0289] Step 9:

[0290] The server sends the updated recommendation list back to the user's device, and the user confirms the new recommendations. The input is the updated recommendation list, and the output is the content displayed on the device. Based on this information, the user can choose their next action.

[0291] Through these processing steps, real-time recommendations based on the user's mood and activity status are provided.

[0292] 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.

[0293] User profile collection and emotion recognition

[0294] 1. A user starts the music streaming application and logs in to begin using the system.

[0295] Example: A user opens an app and enters their login details to access their account.

[0296] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[0297] Example: The app prompts you with questions like, "How are you feeling right now?" or "What are you doing right now?"

[0298] 3. The user enters the information into the terminal, and the terminal collects the information.

[0299] Example: A user enters information such as "I want to relax" or "I'm exercising."

[0300] 4. The emotion engine installed on the device recognizes and collects the user's emotions.

[0301] Example: The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotional data such as "The user looks happy" or "The user is feeling stressed."

[0302] 5. The device retrieves past music playback history from local storage and sends it to the server.

[0303] Example: An app retrieves a list of previously played songs and sends it to a server along with current mood, activity, and emotion data.

[0304] Data analysis and playlist generation

[0305] 1. The server receives the user's current mood, activity, emotion, and playback history data.

[0306] Example: Receive data such as "User ID: 12345, current mood: relaxed, activity: exercising, emotion: happy."

[0307] 2. The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[0308] Example: Create an instance of an AI model and provide profile information such as "relaxed," "exercising," or "happy" as input parameters.

[0309] 3. The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[0310] Example: An AI model selects appropriate songs based on a given profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[0311] 4. The server sends the generated playlist to the user's device.

[0312] Example: The generated playlist is structured in JSON format and sent to the user's device.

[0313] Real-time playlist updates

[0314] 1. The device displays the generated playlist to the user.

[0315] Example: Displaying a list of transmitted playlist data on the screen.

[0316] 2. The user plays a song from the playlist.

[0317] Example: A song selected by the user is played.

[0318] 3. The user enters their feedback on the song into the device, and the emotion engine continuously monitors the user's emotions.

[0319] Example: You input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze your facial expressions and collects emotional data.

[0320] 4. The device sends the user's feedback and emotion data to the server.

[0321] Example: Sending feedback information and emotion data to a server in real time.

[0322] 5. The server regenerates or updates the playlist based on the feedback and sentiment data.

[0323] Example: Analyzing the received data and reselecting the song that best suits the user's current state and emotions.

[0324] 6. The server sends the updated playlist to the user's device.

[0325] Example: New playlist data is structured in JSON format and sent to the user's device for display.

[0326] User interaction

[0327] 1. The user makes a request or asks a question to the AI ​​DJ.

[0328] Example: Enter a question into a form: "What's the next song you recommend?"

[0329] 2. The device sends the user's request or question to the server.

[0330] Example: Sending entered message data to the server.

[0331] 3. The server uses the AI ​​DJ model and emotional data to generate an appropriate response.

[0332] Example: An AI DJ analyzes the request content and emotional information and generates a response message such as, "The next song I recommend to relax you is XX."

[0333] 4. The server generates a response and sends it to the user's device.

[0334] Example: A response message is sent to the user's terminal and displayed.

[0335] 5. The device displays or plays the AI ​​DJ's response to the user.

[0336] Example: The generated response is displayed as text on the user's screen or played in audio format.

[0337] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[0338] The processing flow will be explained below.

[0339] Program processing steps

[0340] ---

[0341] User profile collection and emotion recognition

[0342] Step 1:

[0343] A user launches a music streaming application and logs in.

[0344] What happens: The user enters their username and password on the app's login screen and submits their authentication information.

[0345] Step 2:

[0346] The device asks the user about their current mood and activity.

[0347] What it does: The app prompts you with a pop-up asking questions like "How are you feeling right now?" and "What are you doing right now?"

[0348] Step 3:

[0349] The user inputs their current mood and activity status into the terminal.

[0350] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[0351] Step 4:

[0352] The emotion engine installed in the device recognizes the user's emotions.

[0353] Specific operation: The device's camera takes a picture of the user's face and analyzes their facial expressions. From the analyzed data, emotions such as "happiness" and "stress" are recognized.

[0354] Step 5:

[0355] The device retrieves the user's past music playback history from local storage and transmits it to the server.

[0356] What it does: Reads playback history data stored in local storage and sends it to the server along with current mood, activity, and recognized emotion data.

[0357] ---

[0358] Data analysis and playlist generation

[0359] Step 6:

[0360] The server receives the user's current mood, activity, emotion, and playback history data.

[0361] Specific operation: The server saves the received data in the database and begins analysis.

[0362] Step 7:

[0363] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[0364] What it does: Create an instance of an AI model and provide the user's profile data (e.g., "relaxed," "exercising," "happy") as input parameters.

[0365] Step 8:

[0366] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[0367] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile and emotional data.

[0368] Step 9:

[0369] The server transmits the generated playlist to the user's terminal.

[0370] Specific operation: The generated playlist is structured in JSON format and sent to the user's device.

[0371] ---

[0372] Real-time playlist updates

[0373] Step 10:

[0374] The terminal displays the generated playlist to the user.

[0375] Specific operation: The transmitted playlist data is displayed on the screen as a list.

[0376] Step 11:

[0377] The user plays the songs in the playlist.

[0378] Specific behavior: Starts playing the song selected by the user.

[0379] Step 12:

[0380] The user inputs feedback about the song being played into the terminal, and the emotion engine continues to recognize the user's emotions.

[0381] Specific operation: Input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze the user's facial expressions and collects emotional data.

[0382] Step 13:

[0383] The terminal transmits the user's feedback and emotion data to the server.

[0384] Specific operation: The input feedback information and recognized emotion data are sent to the server in real time.

[0385] Step 14:

[0386] The server regenerates or updates the playlist based on the feedback and emotional data.

[0387] Specific behavior: Analyzes the received data and reselects the song that best suits the user's current situation and emotions.

[0388] Step 15:

[0389] The server sends the updated playlist to the user's terminal.

[0390] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[0391] ---

[0392] User interaction

[0393] Step 16:

[0394] Users make requests and ask questions to the AI ​​DJ.

[0395] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[0396] Step 17:

[0397] The terminal transmits the user's requests and questions to the server.

[0398] Specific operation: Sends the input request and question data to the server.

[0399] Step 18:

[0400] The server uses the AI ​​DJ model and recognized emotion data to generate appropriate responses.

[0401] How it works: The AI ​​DJ model analyzes the request content and emotional data, and generates a response message such as, "The next recommended song is XX."

[0402] Step 19:

[0403] The server generates a response and sends it to the user's terminal.

[0404] Specific operation: Sends a response message to the terminal.

[0405] Step 20:

[0406] The terminal displays or plays the AI ​​DJ's responses to the user.

[0407] What happens: A response message is displayed on the screen as text or played aloud.

[0408] Example 2

[0409] 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."

[0410] While conventional music streaming systems provide playlists that take user preferences into account, they are unable to provide music that instantly adapts to the user's current mood, activity status, or even emotional changes. This makes it difficult to provide the optimal music experience for the user. Furthermore, the lack of a means to adjust the music experience through real-time feedback or dialogue also hinders the quality of the personalized music experience.

[0411] 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.

[0412] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity status, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood, activity status, and emotions, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, and an emotion recognizing means that continuously monitors and recognizes the user's emotions and reflects the acquired emotion data in the music playlist. This makes it possible to instantly respond to the user's preferences, current mood, activity status, and emotional changes and provide an optimal, personalized music experience.

[0413] "User" means any individual or organization that uses this system.

[0414] "Preferences" are information that indicates the user's past choices or usage trends.

[0415] "Mood" is information that indicates the user's current mental or emotional state.

[0416] "Activity status" is information that indicates the specific actions and situations that a user is currently engaged in.

[0417] A "generative artificial intelligence model" is an artificial intelligence algorithm that makes predictions and recommendations based on given data.

[0418] "Music provision means" refers to a function that provides music to users using a generative artificial intelligence model.

[0419] "Interaction means" is a function for communicating with users in real time and updating music playlists.

[0420] The "profile collection means" is a means for acquiring the user's current mood, activity status, and emotions.

[0421] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[0422] "Emotion recognition means" refers to a means for monitoring and recognizing a user's emotions and generating emotion data.

[0423] "Data" refers to various information, including information about a user's mood, activities, preferences, emotions, etc.

[0424] "Feedback" refers to opinions and evaluations provided by users to the system.

[0425] "Real-time" refers to actions or processes that respond immediately and without delay.

[0426] The present invention relates to a system for generating music playlists based on a user's preferences and current mood and activity status, which recognizes and reflects the user's current emotions in the music playlist, and provides a personalized music experience through real-time feedback and interaction.

[0427] User profile collection and emotion recognition

[0428] The user begins using the system by launching a music streaming application and logging in. The device asks the user about their current mood and activity (e.g., studying, exercising, or just wanting to relax). The user enters this information into the device, which then collects it. An emotion engine installed on the device recognizes and collects the user's emotions. The emotion engine uses tools such as Affectiva or Microsoft Emotion API. The device retrieves past music playback history from local storage and sends it to the server.

[0429] Data analysis and playlist generation

[0430] The server receives the user's current mood, activity, emotion, and playback history data. The server loads a generative artificial intelligence model and inputs the user's profile data into the model. This model uses GPT-3 or BERT. The server uses this data to generate a music playlist that best suits the user's current situation and emotion. The generated playlist is structured in JSON format and sent to the user's device.

[0431] Real-time playlist updates

[0432] The device displays the generated playlist to the user. As the user plays songs from the playlist, they can enter their feedback about the songs into the device. The emotion engine continues to monitor the user's emotions and collect emotion data. The device sends the user's feedback and emotion data to the server. The server regenerates or updates the playlist based on the feedback and emotion data. The updated playlist is again sent to the user's device in JSON format and displayed.

[0433] User interaction

[0434] Interaction between the user and the system begins when the user makes a request or question to the AI ​​DJ. For example, the user enters a question into the input form, such as "What song do you recommend next?" The device sends this request or question to the server. The server uses the AI ​​DJ model and emotional data to generate an appropriate response and sends it to the user's device. The device then displays or plays the AI ​​DJ's response to the user.

[0435] Specific examples

[0436] The user opens the app and enters their login details to access their account.

[0437] The app prompts you to enter information by popping up questions such as "How are you feeling right now?" and "What are you doing now?"

[0438] The user inputs information such as "I want to relax" or "I'm exercising."

[0439] The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotion data.

[0440] An instance of the AI ​​model is generated and profile information such as "relaxed," "exercising," and "happy" is provided as input parameters.

[0441] The AI ​​model selects appropriate songs based on the specified profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[0442] The generated playlist is structured in JSON format and sent to the user's device.

[0443] The user types a question into the input form: "What song do you recommend next?"

[0444] The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next song we recommend to help you relax is XX."

[0445] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[0446] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0447] Step 1:

[0448] A user launches a music streaming application and logs in.

[0449] Specific operation: The user taps the app to launch it, and the login screen appears. The user enters their username and password and presses the "Login" button.

[0450] Input: Username, Password

[0451] Output: Login session established, user information obtained

[0452] Step 2:

[0453] The device displays a prompt asking the user about their current mood and activity.

[0454] What it does: The app displays a pop-up window, offering the user options such as "How are you feeling right now?" and "What are you currently doing?", as well as a text input field.

[0455] Input: None

[0456] Output: Questions about mood and activity

[0457] Step 3:

[0458] The user inputs their current mood and activity into the device, and the device collects that information.

[0459] Specific operation: The user enters information such as "I want to relax" or "I'm exercising" and presses the "Send" button. The device saves this information in local storage.

[0460] Input: User's mood, activity information

[0461] Output: Mood and activity information stored in local storage

[0462] Step 4:

[0463] The emotion engine installed in the device recognizes and collects the user's emotions.

[0464] How it works: The device's camera captures the user's face, and the facial expression analysis algorithm generates emotional data such as happiness, sadness, and stress, which is then stored in local storage.

[0465] Input: User's face image

[0466] Output: Parsed emotion data

[0467] Step 5:

[0468] The device retrieves past music playback history from local storage and sends it to the server.

[0469] What it does: The device reads past playback history from local storage and sends it to the server as a JSON-formatted data packet along with current mood, activity, and emotion data.

[0470] Input: Playback history, current mood, activity, and emotion data in local storage

[0471] Output: Data packet sent to the server

[0472] Step 6:

[0473] The server receives the user's current mood, activity, emotion, and playback history data.

[0474] Specific operation: The server stores data packets received via the API endpoint in a database.

[0475] Input: User mood, activity, emotion, and playback history data

[0476] Output: User data stored in the database

[0477] Step 7:

[0478] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[0479] What it does: The server loads an instance of a generative AI model (e.g., GPT-3) into memory and provides the profile data as input parameters.

[0480] Input: User profile data

[0481] Output: Profile data input to the model

[0482] Step 8:

[0483] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[0484] What it does: The model analyzes the input data and lists the best songs. The playlist is structured in JSON format.

[0485] Input: Profile data input to the model

[0486] Output: Structured playlist data

[0487] Step 9:

[0488] The server transmits the generated playlist to the user's terminal.

[0489] Specific operation: Playlist data is sent to the user's device via API, and the device saves the received data.

[0490] Input: Structured playlist data

[0491] Output: Playlist data sent to the user's device

[0492] Step 10:

[0493] The terminal displays the generated playlist to the user.

[0494] Specific operation: Based on the received playlist data, the app displays a song list on the screen.

[0495] Input: Playlist data received from the server

[0496] Output: A list of songs displayed on the screen

[0497] Step 11:

[0498] The user plays a song from the playlist.

[0499] Specific behavior: When the user selects a song and presses the "Play" button, the music will play.

[0500] Input: User song selection

[0501] Output: Played music

[0502] Step 12:

[0503] The user inputs feedback about the song into the terminal, and the emotion engine continuously monitors the user's emotions.

[0504] Specific operation: The user inputs feedback such as "I like this song" or "I want to hear the next song." The device's camera continuously analyzes the user's facial expressions and generates emotion data.

[0505] Input: User feedback, face image

[0506] Output: Feedback data, analyzed emotion data

[0507] Step 13:

[0508] The device transmits the user's feedback and emotional data to the server.

[0509] What it does: Sends feedback and emotion data in real time to a server via an API.

[0510] Input: Feedback data, parsed emotion data

[0511] Output: Feedback and emotion data sent to the server

[0512] Step 14:

[0513] The server regenerates or updates the playlist based on the feedback and emotional data.

[0514] What happens: The server re-runs the AI ​​model based on the new data and generates an updated playlist.

[0515] Input: Feedback and emotion data

[0516] Output: Regenerated or updated playlist data

[0517] Step 15:

[0518] The server sends the updated playlist to the user's terminal.

[0519] Specific behavior: The new playlist data is sent in JSON format to the user's device, and the device displays the playlist again.

[0520] Input: Updated playlist data

[0521] Output: Updated playlist data sent to the user device

[0522] Step 16:

[0523] Users make requests and ask questions to the AI ​​DJ.

[0524] What happens: A user types a question into a text input form, such as "What's the next song you recommend?"

[0525] Input: User request or question

[0526] Output: The input request or question

[0527] Step 17:

[0528] The terminal transmits the user's requests and questions to the server.

[0529] Specific operation: The entered message data is sent to the server via the API.

[0530] Input: Request or question data

[0531] Output: Message data sent to the server

[0532] Step 18:

[0533] The server uses the AI ​​DJ model and emotional data to generate appropriate responses.

[0534] Specific operation: The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next recommended song to help you relax is XX."

[0535] Input: Request or question data, sentiment data

[0536] Output: The generated response message

[0537] Step 19:

[0538] The server generates a response and sends it to the user's terminal.

[0539] Specific operation: A response message is sent to the user's device via the API.

[0540] Input: The generated response message

[0541] Output: Response message sent to the user terminal

[0542] Step 20:

[0543] The terminal displays or plays the AI ​​DJ's responses to the user.

[0544] Specific Behavior: The generated response is displayed to the user in text on their screen or played in audio format.

[0545] Input: Response message sent to the user terminal

[0546] Output: The response message displayed or played on the user's terminal.

[0547] This allows users to get the best possible music experience based on their preferences and emotional information, and allows them to enjoy a more personalized music experience through real-time feedback and interaction.

[0548] (Application example 2)

[0549] 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."

[0550] Conventional music streaming services generate playlists based on users' music preferences, but they do not adequately provide a personalized music experience based on users' real-time emotions and activities. It is also difficult to instantly update playlists to reflect user feedback and emotional data in real time. As a result, it is difficult to provide users with the optimal music experience.

[0551] 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 a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, an emotion recognition means that captures the user's facial expressions and analyzes their emotions, and an emotion input means that inputs the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model. This enables a music experience based on the user's real-time emotions and activities.

[0552] A "music providing means" is a means that includes a generative artificial intelligence model that generates a music playlist based on user preferences.

[0553] The "interaction means" is a means for updating the music playlist in real time by interacting with the user.

[0554] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[0555] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[0556] An "emotion recognition means" is a means for capturing a user's facial expressions and analyzing their emotions.

[0557] The "emotion input means" is a means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[0558] A "generative AI model" is an AI model equipped with an algorithm that generates playlists based on user preferences and updates them based on real-time data.

[0559] A "music playlist" is a list of music generated based on a user's preferences and emotions.

[0560] This invention provides a system that generates a music playlist based on a user's preferences and updates the playlist in real time. The system uses emotional data obtained from the user's current mood, activity status, and facial expressions to provide appropriate music content in a personalized manner. A specific embodiment of this invention will be described below.

[0561] Hardware and software used

[0562] Hardware:

[0563] Webcam: To capture the user's facial expressions

[0564] Smartphones and tablets: To interact with the user and play music

[0565] software:

[0566] OpenCV: A library for capturing and analyzing facial expressions

[0567] TensorFlow: A library for running emotion recognition models

[0568] requests: An HTTP request library for sending data to servers and retrieving music playlists.

[0569] System configuration

[0570] 1. Generative AI model that generates music playlists based on user preferences: Equipped with an AI model that generates optimal playlists based on the music a user has played in the past and their profile information.

[0571] 2. Interaction: A means to interact with the user in real time and update the music playlist. The user makes requests via text or voice input.

[0572] 3. Profile collection means: A means of obtaining a user's current mood and activity status, including information the user enters into the device and daily activity information.

[0573] 4. Input means: A means for inputting data collected by the profile collection means into the generative artificial intelligence model.

[0574] 5. Emotion Recognition: A means of capturing the user's facial expressions and analyzing their emotions. Specifically, facial images taken with a webcam are preprocessed using OpenCV and then analyzed using TensorFlow's emotion recognition model.

[0575] 6. Emotion input means: A means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[0576] Process Overview

[0577] The server first generates a music playlist based on the user's preferences. This process is based on past music playback history, current mood, activity status, and emotional data. The webcam captures the user's facial expressions and analyzes their emotions. This analysis, along with their profile information, is then fed into a generative artificial intelligence model to provide the optimal music playlist.

[0578] Examples of concrete examples and prompts

[0579] Examples:

[0580] The scenario imagines a user wearing smart glasses and a camera capturing their facial expressions.

[0581] The emotion recognition model detects when the user is "happy."

[0582] The server then sends the user a playlist of "relaxing music that matches a happy mood" along with the emotional data.

[0583] Example prompt sentence:

[0584] User ID: user12345, current emotion: happy

[0585] Recommend the best music content for this user.

[0586] This system enables a music experience based on the user's real-time emotions and activities, providing a deeper, more personalized entertainment experience.

[0587] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0588] Step 1:

[0589] A user launches a music streaming application and logs in. The user opens the app, enters their login information, and accesses their account. During this process, a set of initial data is collected, allowing the system to obtain the user's identity and profile information.

[0590] Step 2:

[0591] The device asks the user about their current mood and activity. Specifically, the app displays a pop-up asking questions such as "How are you feeling right now?" and "What are you currently doing?" to prompt the user for input. In response, the user provides information such as "I want to relax" or "I'm exercising." This information is collected by the profile collection means.

[0592] Step 3:

[0593] The device uses a webcam to capture the user's facial expressions. The emotion recognition unit preprocesses this facial image and converts it to grayscale. It then uses a face detection model to identify the face's location. The identified facial features are input into TensorFlow's emotion recognition model to analyze the user's emotions. The analysis results in emotional data such as "happiness," "anger," or "surprise."

[0594] Step 4:

[0595] The device sends the collected mood and activity data, as well as the analyzed emotional data, to the server, which then structures the data along with the profile information and sends it to the server, which receives it and prepares the data for real-time analysis.

[0596] Step 5:

[0597] The server inputs the received data into a generative AI model. The model is fed with the user's current mood, activity, emotions, and past music playback history. The generative AI model generates an optimal music playlist based on this data. The model analyzes the data and selects music to generate an optimal song list.

[0598] Step 6:

[0599] The server sends the generated music playlist to the device. The playlist is structured in JSON format and sent to the user's device. The device receives the playlist and displays it to the user.

[0600] Step 7:

[0601] The user plays songs from the generated playlist, and the device plays the selected songs and collects user feedback during the process. Feedback information is collected through simple form-filling and sentiment analysis.

[0602] Step 8:

[0603] The device sends the collected feedback and real-time emotional data to the server, which then inputs it back into the generative AI model to regenerate or update the playlist. The updated playlist is then sent back to the device and displayed to the user. This cycle creates a dynamic music experience for the user.

[0604] 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.

[0605] 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.

[0606] 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.

[0607] [Second embodiment]

[0608] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0609] 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.

[0610] 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).

[0611] 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.

[0612] 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.

[0613] 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).

[0614] 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.

[0615] 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.

[0616] 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.

[0617] 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.

[0618] 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.

[0619] 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."

[0620] Collecting user profiles

[0621] 1. A user starts the music streaming application and begins using the system by logging in.

[0622] Example: A user opens an app and enters their login details to access their account.

[0623] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[0624] Example: An app pops up a question to the user, such as "How are you feeling right now?" or "What are your current activities?"

[0625] 3. The user enters the information into the terminal.

[0626] Example: A user enters information such as "I want to relax" or "I'm exercising."

[0627] 4. The device retrieves past music playback history from local storage and sends it to the server.

[0628] Example: An app gets a list of previously played songs and sends it to a server.

[0629] Data analysis and playlist generation

[0630] 1. The server receives the user's preferences and current state data.

[0631] Example: Receiving data such as "User ID: 12345, current mood: relaxed, activity: exercising."

[0632] 2. The server loads the generative artificial intelligence model and inputs the user profile.

[0633] Example: Providing profile information such as "relaxed" or "exercising" to an AI model.

[0634] 3. The server uses this data to generate a music playlist that best suits the user's current situation.

[0635] Example: Generate a playlist of relaxing and exercise-friendly music such as "Lo-fi Hip Hop" and "House Music."

[0636] 4. The server sends the generated playlist to the user's device.

[0637] Example: The generated playlist is sent to the app and displayed to the user.

[0638] Real-time playlist updates

[0639] 1. Listen to songs from a user-provided playlist.

[0640] Example: A user starts playing a song from a recommended playlist.

[0641] 2. The user inputs feedback about the song being played into the device, such as whether they like it or not, or if they want to skip it.

[0642] Example: Enter feedback such as "I like this song" or "I want to hear the next song."

[0643] 3. The device sends the feedback to the server.

[0644] Example: An app sends feedback information to a server in real time.

[0645] 4. The server regenerates or updates the playlist based on the feedback.

[0646] Example: Adding or changing songs in a playlist based on new feedback.

[0647] 5. The server sends the updated playlist back to the user's device.

[0648] Example: The updated playlist is reflected and displayed on the user's device.

[0649] User interaction

[0650] 1. The user makes a request or asks the AI ​​DJ a question, such as "What song do you recommend next?"

[0651] Example: A user types the question "What song should I play next?"

[0652] 2. The device sends the user's request or question to the server.

[0653] Example: An app sends a message from the user to a server.

[0654] 3. The server uses the AI ​​DJ model to generate an appropriate response.

[0655] Example: An AI DJ generates messages such as, "The song I recommend for your next relaxation session is XX."

[0656] 4. The server generates a response and sends it to the user's device.

[0657] Example: A response message is displayed in the user's app.

[0658] 5. The device displays or plays the AI ​​DJ's response to the user.

[0659] For example: The generated response is displayed or played in text or audio format.

[0660] This system allows users to get the best possible music experience based on their preferences and current situation, and allows for a more personalized music experience through real-time feedback and interaction.

[0661] The processing flow will be explained below.

[0662] Program processing steps

[0663] ---

[0664] Collecting user profiles

[0665] Step 1:

[0666] A user launches a music streaming application and logs in.

[0667] Specific behavior: Enter your username and password on the app's login screen and submit your authentication information.

[0668] Step 2:

[0669] The device asks the user about their current mood and activity.

[0670] Specific behavior: The app prompts the user with questions such as "How are you feeling right now?" and "What are you currently doing?"

[0671] Step 3:

[0672] The user inputs their current mood and activity status into the terminal.

[0673] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[0674] Step 4:

[0675] The device retrieves past music playback history from local storage and sends it to the server.

[0676] Specific operation: Reads playback history data from local storage and sends it to the server along with collected mood and activity information.

[0677] ---

[0678] Data analysis and playlist generation

[0679] Step 5:

[0680] The server receives the user's current mood, activity, and playback history data.

[0681] Specific operation: The server saves the received data in the database and begins analysis.

[0682] Step 6:

[0683] The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[0684] Specific operation: Create an instance of the AI ​​model and pass profile data such as "relaxed" or "exercising" as input parameters.

[0685] Step 7:

[0686] The server uses an AI model to generate a playlist based on the user's profile.

[0687] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile.

[0688] Step 8:

[0689] The server transmits the generated playlist to the user's terminal.

[0690] Specific operation: The selected song list is structured in JSON format and sent to the user's device.

[0691] ---

[0692] Real-time playlist updates

[0693] Step 9:

[0694] The terminal displays the generated playlist to the user.

[0695] Specific behavior: Parse the transmitted playlist data and display it in the app interface.

[0696] Step 10:

[0697] The user plays a song from the playlist.

[0698] Specific behavior: The song selected by the user will be played within the app.

[0699] Step 11:

[0700] The user inputs feedback about the song into the terminal.

[0701] Specific action: Enter feedback such as "I like this song" or "I want to hear the next song" into the input form.

[0702] Step 12:

[0703] The terminal transmits the feedback from the user to the server.

[0704] Specific behavior: The input feedback is sent to the server in real time.

[0705] Step 13:

[0706] The server regenerates or updates the playlist based on the feedback.

[0707] Specific behavior: Analyzes the received feedback data and performs processing to update the playlist.

[0708] Step 14:

[0709] The server sends the updated playlist to the user's terminal.

[0710] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[0711] ---

[0712] User interaction

[0713] Step 15:

[0714] Users make requests and ask questions to the AI ​​DJ.

[0715] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[0716] Step 16:

[0717] The terminal transmits the user's requests and questions to the server.

[0718] Specific operation: Sends the entered message data to the server.

[0719] Step 17:

[0720] The server uses an AI DJ model to generate appropriate responses.

[0721] Specific operation: The AI ​​DJ model analyzes the request content and generates an appropriate response message.

[0722] Step 18:

[0723] The server generates a response and sends it to the user's terminal.

[0724] Specific operation: The generated response message is sent to the user's terminal.

[0725] Step 19:

[0726] The terminal displays or plays the AI ​​DJ's responses to the user.

[0727] Specific behavior: The response message sent is displayed on the screen or played aloud.

[0728] Example 1

[0729] 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."

[0730] Conventional music streaming systems have struggled to provide real-time music playlists that respond to users' preferences, current moods, and activity status. Furthermore, they lack the ability to respond immediately to user feedback and requests, making it impossible to provide an optimized music experience for each individual user. The present invention aims to solve these problems and provide users with a more personalized music experience.

[0731] 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.

[0732] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model. This makes it possible to generate and update an optimal music playlist in real time based on the user's preferences and current status and provide it to the user.

[0733] "User" refers to an individual who uses the music streaming system.

[0734] "Preferences" refers to a user's musical tastes and tendencies.

[0735] "Mood" refers to the emotion or mental state a user is feeling at a particular time.

[0736] An "activity" is an action or task that a user is performing at a particular time.

[0737] "Music Playlist" means a list of songs organized for sequential playback by a user.

[0738] A "generative artificial intelligence model" refers to an artificial intelligence technology that generates new information or recommendations based on input data.

[0739] "Music provision means" refers to a mechanism that uses a generative artificial intelligence model to provide users with music playlists.

[0740] "Interaction means" refers to a mechanism by which users and systems communicate with each other.

[0741] "Profile Collection Measures" refers to mechanisms that collect information about a user's preferences, moods, and activities.

[0742] "Input means" refers to a mechanism for inputting data collected by the profile collection means into the generative artificial intelligence model.

[0743] "Server" refers to the computer system that processes user data and runs the generative artificial intelligence model.

[0744] This invention relates to a system that generates and updates an optimal music playlist in real time according to a user's preferences, current mood, and activity status. Specific embodiments of this system are described in detail below.

[0745] The system is mainly composed of three entities: a server, a terminal, and a user.

[0746] Collecting user profiles

[0747] A user begins using the system by launching a music streaming application and logging in. The user opens the app on a device such as a smartphone or tablet and enters their login information for authentication. If authentication is successful, the device displays a pop-up window asking about the user's current mood and activity. When the user enters information such as "I want to relax" or "I'm exercising," the device collects this information and retrieves past music playback history from local storage.

[0748] Data analysis and playlist generation

[0749] The data collected by the device is sent to the server. The server loads a generative artificial intelligence model and inputs the user profile based on the received user preferences and current state data. The generative AI model uses this data to generate a music playlist that is optimal for the user's current situation. The generated playlist is then sent back to the user's device, where it is displayed.

[0750] Real-time playlist updates

[0751] While listening to songs in a playlist, a user can input feedback about the song being played into the device. If the user inputs feedback such as "I like this song" or "I want to hear the next song," the device sends this information to the server in real time. The server regenerates or updates the playlist based on this feedback and sends the new playlist to the user's device.

[0752] User interaction

[0753] Users can make requests or questions to the AI ​​DJ. For example, if they input a question like "What's the next song you recommend?", the device will send this request to the server. The server will use the AI ​​DJ model to generate an appropriate response and send it to the user's device. The device will then display or play this response in text or audio format.

[0754] The system allows users to get the best possible music experience in real time based on their preferences and current situation, and further personalizes the music experience through user feedback and interaction.

[0755] Examples of prompt statements

[0756] Prompt: "What are the steps for a system that allows a user to enter their current mood and activity in a music streaming app and have AI generate the perfect playlist?"

[0757] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0758] Step 1:

[0759] The user launches the app and logs in

[0760] A user launches a music streaming app on a device such as a smartphone or tablet and enters their login information (email address, password). The device sends this input data to the server, which then authenticates the user. If authentication is successful, the server returns an authentication success message to the device, allowing the user to access their account.

[0761] Input: Email address, password

[0762] Output: Authentication success message

[0763] Step 2:

[0764] The device asks about the user's mood and activity

[0765] After logging in, the device displays a pop-up window asking the user questions such as "How are you feeling now?" and "What is your current activity?" To make it easier for the user to answer, options (e.g., relaxing, exercising, etc.) are displayed. These options may also be provided by the server.

[0766] Input: Authentication success message

[0767] Output: Mood and activity question popup

[0768] Step 3:

[0769] The user enters information

[0770] The user selects and inputs their mood and activity (e.g., "I want to relax" or "I'm exercising") from a pop-up window on the device, and the device immediately sends this information to the server.

[0771] Input: Questions about mood and activity

[0772] Output: User's mood and activity information

[0773] Step 4:

[0774] The device acquires past music playback history and sends it to the server.

[0775] The device searches for and retrieves past music playback history from its internal local storage, and the retrieved data is sent to the server.

[0776] Input: User's mood and activity information

[0777] Output: Past music playback history

[0778] Step 5:

[0779] The server receives the user's preferences and current state data.

[0780] The server receives data packets from the device about the user's mood, activity, and playback history, stores them in a database, and prepares them for analysis.

[0781] Input: User's mood and activity information, past music playback history

[0782] Output: Data stored in the database

[0783] Step 6:

[0784] The server loads the generative artificial intelligence model and inputs the user profile.

[0785] The server loads the generative AI model file from disk and expands it into memory. Data on the user's current mood, activity, and playback history is input into the generative AI model, and analysis begins.

[0786] Input: Data stored in a database

[0787] Output: User profile fed into the AI ​​model

[0788] Step 7:

[0789] The server generates a music playlist that best suits the user's current situation.

[0790] The server uses an AI model to select the best songs based on the user's input data and generate a playlist, which is then packaged into packets and sent to the device.

[0791] Input: User profile fed into the AI ​​model

[0792] Output: Generated music playlist

[0793] Step 8:

[0794] The server sends the generated playlist to the device.

[0795] The server assembles the generated playlist into a data packet and sends it to the user's terminal, which receives it and displays it to the user.

[0796] Input: Generated music playlist

[0797] Output: Playlist displayed on device

[0798] Step 9:

[0799] Listen to songs from a user-provided playlist

[0800] The user presses the play button on the playlist screen within the app to begin listening to the song, and the device responds by streaming the song data from the server.

[0801] Input: Playlist displayed on device

[0802] Output: The song being played

[0803] Step 10:

[0804] Users provide feedback on songs

[0805] Users can input feedback by tapping buttons on the playback screen such as "I like this song" or "I want to listen to the next song." The device then sends this feedback information to the server in real time.

[0806] Input: User feedback

[0807] Output: Feedback information sent to the server

[0808] Step 11:

[0809] The server regenerates or updates the playlist based on the feedback

[0810] The server re-runs the AI ​​model based on the received feedback, updates the playlist, and reconstructs the newly generated playlist and sends it to the device.

[0811] Input: Feedback information sent to the server

[0812] Output: Regenerated music playlist

[0813] Step 12:

[0814] The server sends the updated playlist to the device.

[0815] The server packages the updated playlist into a data packet and sends it to the user's terminal, which receives it and displays the new playlist.

[0816] Input: Regenerated music playlist

[0817] Output: Updated playlist displayed on device

[0818] Step 13:

[0819] Users make requests and ask questions to the AI ​​DJ

[0820] Users can type or speak using the in-app chat box or voice assistant function, asking, "What's your recommendation for the next song?"

[0821] Input: User requests or questions

[0822] Output: Requests and questions typed into the terminal

[0823] Step 14:

[0824] The device sends requests and questions to the server

[0825] The device sends the user's request or question as a data packet to the server, which receives the data and begins analyzing it.

[0826] Input: Requests or questions typed into the device

[0827] Output: The request or question sent to the server

[0828] Step 15:

[0829] The server generates a response using the AI ​​DJ model

[0830] The server runs the AI ​​DJ model to generate appropriate responses to requests and questions, which are then packaged into data packets and sent to the device.

[0831] Input: The request or question sent to the server

[0832] Output: The generated response message

[0833] Step 16:

[0834] Sends the server-generated response to the terminal

[0835] The server generates a response message and sends it as a data packet to the terminal, which receives it and displays or plays it audibly to the user.

[0836] Input: The generated response message

[0837] Output: The greeting message displayed or played on the terminal

[0838] (Application example 1)

[0839] 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."

[0840] Conventional food delivery systems have struggled to recommend the best food and restaurant in real time based on a user's preferences and current mood. Furthermore, they lacked the technology to instantly update recommendations based on user feedback. This resulted in a lack of consistency and individualization in the user experience.

[0841] 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.

[0842] In this invention, the server includes a recommendation means equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences, an interaction means that updates the recommendations in real time by interacting with the user, a profile collection means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collection means into the generative artificial intelligence model. This makes it possible to recommend optimal dishes and restaurants in real time based on the user's mood and activity status, and to instantly update the recommendations in response to feedback.

[0843] A "recommendation vehicle" is a vehicle equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences.

[0844] An "interaction means" is a means for updating recommendations in real time by interacting with the user.

[0845] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[0846] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[0847] A "generative artificial intelligence model" is an artificial intelligence model that generates optimal recommendations based on user preferences.

[0848] "Past order history" refers to the history of orders previously placed by the user.

[0849] An "algorithm" is a set of steps or rules for solving a specific problem by following a set of steps.

[0850] The system for implementing this invention mainly comprises a user terminal, a server, and a generative artificial intelligence model. The user terminal includes a mobile device such as a smartphone, and the server is located in a remote data center. The system has the function of collecting user profile information and providing optimal recommendations based on that information.

[0851] First, when a user logs in to the application using their smartphone, the device asks about the user's current mood and activity status. When the user enters information such as "I want to relax" or "I'm at a party," the device retrieves this information from local storage and sends it to the server.

[0852] The server receives the user's current status and past order history. Specifically, data such as "User ID: 12345, current mood: relaxed, activity: partying" is received and input into a generative AI model. This generative AI model generates optimal recommendations based on the user profile.

[0853] The generative AI model then uses the user's profile information to generate a list of recommended dishes and restaurants. For example, if a user wants to relax, snacks and desserts might be recommended, while if they're at a party, pizza and alcoholic drinks might be recommended.

[0854] Once the recommendation list is generated, the server sends the information to the user's device and displays it on the device. The user selects from the recommendation list and enters feedback into the device. For example, the user may enter feedback such as "I like this dish" or "I'd like to try a different dish next time." The device then sends this feedback to the server in real time, and the server again uses the generative artificial intelligence model to update the recommendation list.

[0855] The hardware used includes smartphones (user devices) and servers (remote data centers), and the software includes Python, Flask, generative artificial intelligence model libraries (e.g., some_ai_lib), JSON, HTTP, etc.

[0856] For example, if the prompt information "Current opinion: Relax, Activity: Partying" is entered, the generative artificial intelligence model will provide a recommendation based on this prompt information, such as "The next recommended dish is sushi and a tapioca drink."

[0857] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0858] Step 1:

[0859] A user logs in to an application on a smartphone. The user enters login information and accesses the application. The device receives this login information and performs user authentication. The input is the login information, and the output is the authentication status (success or failure).

[0860] Step 2:

[0861] If authentication is successful, the device displays a pop-up asking the user about their current mood and activity status. The user inputs information such as "I want to relax" or "I'm at a party." The input is the user's mood and activity status, and the output is a data format containing this information.

[0862] Step 3:

[0863] The device retrieves the user's current mood and activity status from local storage and sends it to the server. The input is the data in local storage and newly entered information, and the output is the data format (e.g., JSON format) sent to the server.

[0864] Step 4:

[0865] The server receives the user's current mood, activity status, and past order history and inputs them into a generative artificial intelligence model. The inputs are the user's current mood, activity status, and past order history, and the output is the profile data that is fed into the model.

[0866] Step 5:

[0867] A generative artificial intelligence model uses a user's profile information to generate a list of recommendations for the most suitable dishes and restaurants. The input is the user profile information, and the output is a list of recommendations. Specifically, the AI ​​model processes the data by analyzing past history and current status, and then uses a matching algorithm to identify recommended items.

[0868] Step 6:

[0869] The server sends the generated recommendation list to the user terminal. The input is the generated recommendation list, and the output is the recommendation list displayed on the user terminal. The user terminal displays this list on its screen.

[0870] Step 7:

[0871] The user selects from the recommendation list and inputs feedback into the terminal, for example, "I like this dish" or "I'd like to try a different dish next time." The input is the user's feedback, and the output is the feedback data sent to the server.

[0872] Step 8:

[0873] The device sends real-time feedback to the server, which then uses the generative AI model to update the recommendation list. The input is the feedback data, and the output is an updated recommendation list. Specifically, the AI ​​model re-runs optimization based on the new feedback to calculate the optimal recommendations.

[0874] Step 9:

[0875] The server sends the updated recommendation list back to the user's device, and the user confirms the new recommendations. The input is the updated recommendation list, and the output is the content displayed on the device. Based on this information, the user can choose their next action.

[0876] Through these processing steps, real-time recommendations based on the user's mood and activity status are provided.

[0877] 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.

[0878] User profile collection and emotion recognition

[0879] 1. A user starts the music streaming application and logs in to begin using the system.

[0880] Example: A user opens an app and enters their login details to access their account.

[0881] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[0882] Example: The app prompts you with questions like, "How are you feeling right now?" or "What are you doing right now?"

[0883] 3. The user enters the information into the terminal, and the terminal collects the information.

[0884] Example: A user enters information such as "I want to relax" or "I'm exercising."

[0885] 4. The emotion engine installed on the device recognizes and collects the user's emotions.

[0886] Example: The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotional data such as "The user looks happy" or "The user is feeling stressed."

[0887] 5. The device retrieves past music playback history from local storage and sends it to the server.

[0888] Example: An app retrieves a list of previously played songs and sends it to a server along with current mood, activity, and emotion data.

[0889] Data analysis and playlist generation

[0890] 1. The server receives the user's current mood, activity, emotion, and playback history data.

[0891] Example: Receive data such as "User ID: 12345, current mood: relaxed, activity: exercising, emotion: happy."

[0892] 2. The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[0893] Example: Create an instance of an AI model and provide profile information such as "relaxed," "exercising," or "happy" as input parameters.

[0894] 3. The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[0895] Example: An AI model selects appropriate songs based on a given profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[0896] 4. The server sends the generated playlist to the user's device.

[0897] Example: The generated playlist is structured in JSON format and sent to the user's device.

[0898] Real-time playlist updates

[0899] 1. The device displays the generated playlist to the user.

[0900] Example: Displaying a list of transmitted playlist data on the screen.

[0901] 2. The user plays a song from the playlist.

[0902] Example: A song selected by the user is played.

[0903] 3. The user enters their feedback on the song into the device, and the emotion engine continuously monitors the user's emotions.

[0904] Example: You input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze your facial expressions and collects emotional data.

[0905] 4. The device sends the user's feedback and emotion data to the server.

[0906] Example: Sending feedback information and emotion data to a server in real time.

[0907] 5. The server regenerates or updates the playlist based on the feedback and sentiment data.

[0908] Example: Analyzing the received data and reselecting the song that best suits the user's current state and emotions.

[0909] 6. The server sends the updated playlist to the user's device.

[0910] Example: New playlist data is structured in JSON format and sent to the user's device for display.

[0911] User interaction

[0912] 1. The user makes a request or asks a question to the AI ​​DJ.

[0913] Example: Enter a question into a form: "What's the next song you recommend?"

[0914] 2. The device sends the user's request or question to the server.

[0915] Example: Sending entered message data to the server.

[0916] 3. The server uses the AI ​​DJ model and emotional data to generate an appropriate response.

[0917] Example: An AI DJ analyzes the request content and emotional information and generates a response message such as, "The next song I recommend to relax you is XX."

[0918] 4. The server generates a response and sends it to the user's device.

[0919] Example: A response message is sent to the user's terminal and displayed.

[0920] 5. The device displays or plays the AI ​​DJ's response to the user.

[0921] Example: The generated response is displayed as text on the user's screen or played in audio format.

[0922] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[0923] The processing flow will be explained below.

[0924] Program processing steps

[0925] ---

[0926] User profile collection and emotion recognition

[0927] Step 1:

[0928] A user launches a music streaming application and logs in.

[0929] What happens: The user enters their username and password on the app's login screen and submits their authentication information.

[0930] Step 2:

[0931] The device asks the user about their current mood and activity.

[0932] What it does: The app prompts you with a pop-up asking questions like "How are you feeling right now?" and "What are you doing right now?"

[0933] Step 3:

[0934] The user inputs their current mood and activity status into the terminal.

[0935] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[0936] Step 4:

[0937] The emotion engine installed in the device recognizes the user's emotions.

[0938] Specific operation: The device's camera takes a picture of the user's face and analyzes their facial expressions. From the analyzed data, emotions such as "happiness" and "stress" are recognized.

[0939] Step 5:

[0940] The device retrieves the user's past music playback history from local storage and transmits it to the server.

[0941] What it does: Reads playback history data stored in local storage and sends it to the server along with current mood, activity, and recognized emotion data.

[0942] ---

[0943] Data analysis and playlist generation

[0944] Step 6:

[0945] The server receives the user's current mood, activity, emotion, and playback history data.

[0946] Specific operation: The server saves the received data in the database and begins analysis.

[0947] Step 7:

[0948] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[0949] What it does: Create an instance of an AI model and provide the user's profile data (e.g., "relaxed," "exercising," "happy") as input parameters.

[0950] Step 8:

[0951] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[0952] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile and emotional data.

[0953] Step 9:

[0954] The server transmits the generated playlist to the user's terminal.

[0955] Specific operation: The generated playlist is structured in JSON format and sent to the user's device.

[0956] ---

[0957] Real-time playlist updates

[0958] Step 10:

[0959] The terminal displays the generated playlist to the user.

[0960] Specific operation: The transmitted playlist data is displayed on the screen as a list.

[0961] Step 11:

[0962] The user plays the songs in the playlist.

[0963] Specific behavior: Starts playing the song selected by the user.

[0964] Step 12:

[0965] The user inputs feedback about the song being played into the terminal, and the emotion engine continues to recognize the user's emotions.

[0966] Specific operation: Input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze the user's facial expressions and collects emotional data.

[0967] Step 13:

[0968] The terminal transmits the user's feedback and emotion data to the server.

[0969] Specific operation: The input feedback information and recognized emotion data are sent to the server in real time.

[0970] Step 14:

[0971] The server regenerates or updates the playlist based on the feedback and emotional data.

[0972] Specific behavior: Analyzes the received data and reselects the song that best suits the user's current situation and emotions.

[0973] Step 15:

[0974] The server sends the updated playlist to the user's terminal.

[0975] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[0976] ---

[0977] User interaction

[0978] Step 16:

[0979] Users make requests and ask questions to the AI ​​DJ.

[0980] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[0981] Step 17:

[0982] The terminal transmits the user's requests and questions to the server.

[0983] Specific operation: Sends the input request and question data to the server.

[0984] Step 18:

[0985] The server uses the AI ​​DJ model and recognized emotion data to generate appropriate responses.

[0986] How it works: The AI ​​DJ model analyzes the request content and emotional data, and generates a response message such as, "The next recommended song is XX."

[0987] Step 19:

[0988] The server generates a response and sends it to the user's terminal.

[0989] Specific operation: Sends a response message to the terminal.

[0990] Step 20:

[0991] The terminal displays or plays the AI ​​DJ's responses to the user.

[0992] What happens: A response message is displayed on the screen as text or played aloud.

[0993] Example 2

[0994] 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."

[0995] While conventional music streaming systems provide playlists that take user preferences into account, they are unable to provide music that instantly adapts to the user's current mood, activity status, or even emotional changes. This makes it difficult to provide the optimal music experience for the user. Furthermore, the lack of a means to adjust the music experience through real-time feedback or dialogue also hinders the quality of the personalized music experience.

[0996] 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.

[0997] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity status, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood, activity status, and emotions, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, and an emotion recognizing means that continuously monitors and recognizes the user's emotions and reflects the acquired emotion data in the music playlist. This makes it possible to instantly respond to the user's preferences, current mood, activity status, and emotional changes and provide an optimal, personalized music experience.

[0998] "User" means any individual or organization that uses this system.

[0999] "Preferences" are information that indicates the user's past choices or usage trends.

[1000] "Mood" is information that indicates the user's current mental or emotional state.

[1001] "Activity status" is information that indicates the specific actions and situations that a user is currently engaged in.

[1002] A "generative artificial intelligence model" is an artificial intelligence algorithm that makes predictions and recommendations based on given data.

[1003] "Music provision means" refers to a function that provides music to users using a generative artificial intelligence model.

[1004] "Interaction means" is a function for communicating with users in real time and updating music playlists.

[1005] The "profile collection means" is a means for acquiring the user's current mood, activity status, and emotions.

[1006] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[1007] "Emotion recognition means" refers to a means for monitoring and recognizing a user's emotions and generating emotion data.

[1008] "Data" refers to various information, including information about a user's mood, activities, preferences, emotions, etc.

[1009] "Feedback" refers to opinions and evaluations provided by users to the system.

[1010] "Real-time" refers to actions or processes that respond immediately and without delay.

[1011] The present invention relates to a system for generating music playlists based on a user's preferences and current mood and activity status, which recognizes and reflects the user's current emotions in the music playlist, and provides a personalized music experience through real-time feedback and interaction.

[1012] User profile collection and emotion recognition

[1013] The user begins using the system by launching a music streaming application and logging in. The device asks the user about their current mood and activity (e.g., studying, exercising, or just wanting to relax). The user enters this information into the device, which then collects it. An emotion engine installed on the device recognizes and collects the user's emotions. The emotion engine uses tools such as Affectiva or Microsoft Emotion API. The device retrieves past music playback history from local storage and sends it to the server.

[1014] Data analysis and playlist generation

[1015] The server receives the user's current mood, activity, emotion, and playback history data. The server loads a generative artificial intelligence model and inputs the user's profile data into the model. This model uses GPT-3 or BERT. The server uses this data to generate a music playlist that best suits the user's current situation and emotion. The generated playlist is structured in JSON format and sent to the user's device.

[1016] Real-time playlist updates

[1017] The device displays the generated playlist to the user. As the user plays songs from the playlist, they can enter their feedback about the songs into the device. The emotion engine continues to monitor the user's emotions and collect emotion data. The device sends the user's feedback and emotion data to the server. The server regenerates or updates the playlist based on the feedback and emotion data. The updated playlist is again sent to the user's device in JSON format and displayed.

[1018] User interaction

[1019] Interaction between the user and the system begins when the user makes a request or question to the AI ​​DJ. For example, the user enters a question into the input form, such as "What song do you recommend next?" The device sends this request or question to the server. The server uses the AI ​​DJ model and emotional data to generate an appropriate response and sends it to the user's device. The device then displays or plays the AI ​​DJ's response to the user.

[1020] Specific examples

[1021] The user opens the app and enters their login details to access their account.

[1022] The app prompts you to enter information by popping up questions such as "How are you feeling right now?" and "What are you doing now?"

[1023] The user inputs information such as "I want to relax" or "I'm exercising."

[1024] The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotion data.

[1025] An instance of the AI ​​model is generated and profile information such as "relaxed," "exercising," and "happy" is provided as input parameters.

[1026] The AI ​​model selects appropriate songs based on the specified profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[1027] The generated playlist is structured in JSON format and sent to the user's device.

[1028] The user types a question into the input form: "What song do you recommend next?"

[1029] The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next song we recommend to help you relax is XX."

[1030] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[1031] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1032] Step 1:

[1033] A user launches a music streaming application and logs in.

[1034] Specific operation: The user taps the app to launch it, and the login screen appears. The user enters their username and password and presses the "Login" button.

[1035] Input: Username, Password

[1036] Output: Login session established, user information obtained

[1037] Step 2:

[1038] The device displays a prompt asking the user about their current mood and activity.

[1039] What it does: The app displays a pop-up window, offering the user options such as "How are you feeling right now?" and "What are you currently doing?", as well as a text input field.

[1040] Input: None

[1041] Output: Questions about mood and activity

[1042] Step 3:

[1043] The user inputs their current mood and activity into the device, and the device collects that information.

[1044] Specific operation: The user enters information such as "I want to relax" or "I'm exercising" and presses the "Send" button. The device saves this information in local storage.

[1045] Input: User's mood, activity information

[1046] Output: Mood and activity information stored in local storage

[1047] Step 4:

[1048] The emotion engine installed in the device recognizes and collects the user's emotions.

[1049] How it works: The device's camera captures the user's face, and the facial expression analysis algorithm generates emotional data such as happiness, sadness, and stress, which is then stored in local storage.

[1050] Input: User's face image

[1051] Output: Parsed emotion data

[1052] Step 5:

[1053] The device retrieves past music playback history from local storage and sends it to the server.

[1054] What it does: The device reads past playback history from local storage and sends it to the server as a JSON-formatted data packet along with current mood, activity, and emotion data.

[1055] Input: Playback history, current mood, activity, and emotion data in local storage

[1056] Output: Data packet sent to the server

[1057] Step 6:

[1058] The server receives the user's current mood, activity, emotion, and playback history data.

[1059] Specific operation: The server stores data packets received via the API endpoint in a database.

[1060] Input: User mood, activity, emotion, and playback history data

[1061] Output: User data stored in the database

[1062] Step 7:

[1063] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[1064] What it does: The server loads an instance of a generative AI model (e.g., GPT-3) into memory and provides the profile data as input parameters.

[1065] Input: User profile data

[1066] Output: Profile data input to the model

[1067] Step 8:

[1068] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[1069] What it does: The model analyzes the input data and lists the best songs. The playlist is structured in JSON format.

[1070] Input: Profile data input to the model

[1071] Output: Structured playlist data

[1072] Step 9:

[1073] The server transmits the generated playlist to the user's terminal.

[1074] Specific operation: Playlist data is sent to the user's device via API, and the device saves the received data.

[1075] Input: Structured playlist data

[1076] Output: Playlist data sent to the user's device

[1077] Step 10:

[1078] The terminal displays the generated playlist to the user.

[1079] Specific operation: Based on the received playlist data, the app displays a song list on the screen.

[1080] Input: Playlist data received from the server

[1081] Output: A list of songs displayed on the screen

[1082] Step 11:

[1083] The user plays a song from the playlist.

[1084] Specific behavior: When the user selects a song and presses the "Play" button, the music will play.

[1085] Input: User song selection

[1086] Output: Played music

[1087] Step 12:

[1088] The user inputs feedback about the song into the terminal, and the emotion engine continuously monitors the user's emotions.

[1089] Specific operation: The user inputs feedback such as "I like this song" or "I want to hear the next song." The device's camera continuously analyzes the user's facial expressions and generates emotion data.

[1090] Input: User feedback, face image

[1091] Output: Feedback data, analyzed emotion data

[1092] Step 13:

[1093] The device transmits the user's feedback and emotional data to the server.

[1094] What it does: Sends feedback and emotion data in real time to a server via an API.

[1095] Input: Feedback data, parsed emotion data

[1096] Output: Feedback and emotion data sent to the server

[1097] Step 14:

[1098] The server regenerates or updates the playlist based on the feedback and emotional data.

[1099] What happens: The server re-runs the AI ​​model based on the new data and generates an updated playlist.

[1100] Input: Feedback and emotion data

[1101] Output: Regenerated or updated playlist data

[1102] Step 15:

[1103] The server sends the updated playlist to the user's terminal.

[1104] Specific behavior: The new playlist data is sent in JSON format to the user's device, and the device displays the playlist again.

[1105] Input: Updated playlist data

[1106] Output: Updated playlist data sent to the user device

[1107] Step 16:

[1108] Users make requests and ask questions to the AI ​​DJ.

[1109] What happens: A user types a question into a text input form, such as "What's the next song you recommend?"

[1110] Input: User request or question

[1111] Output: The input request or question

[1112] Step 17:

[1113] The terminal transmits the user's requests and questions to the server.

[1114] Specific operation: The entered message data is sent to the server via the API.

[1115] Input: Request or question data

[1116] Output: Message data sent to the server

[1117] Step 18:

[1118] The server uses the AI ​​DJ model and emotional data to generate appropriate responses.

[1119] Specific operation: The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next recommended song to help you relax is XX."

[1120] Input: Request or question data, sentiment data

[1121] Output: The generated response message

[1122] Step 19:

[1123] The server generates a response and sends it to the user's terminal.

[1124] Specific operation: A response message is sent to the user's device via the API.

[1125] Input: The generated response message

[1126] Output: Response message sent to the user terminal

[1127] Step 20:

[1128] The terminal displays or plays the AI ​​DJ's responses to the user.

[1129] Specific Behavior: The generated response is displayed to the user in text on their screen or played in audio format.

[1130] Input: Response message sent to the user terminal

[1131] Output: The response message displayed or played on the user's terminal.

[1132] This allows users to get the best possible music experience based on their preferences and emotional information, and allows them to enjoy a more personalized music experience through real-time feedback and interaction.

[1133] (Application example 2)

[1134] 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."

[1135] Conventional music streaming services generate playlists based on users' music preferences, but they do not adequately provide a personalized music experience based on users' real-time emotions and activities. It is also difficult to instantly update playlists to reflect user feedback and emotional data in real time. As a result, it is difficult to provide users with the optimal music experience.

[1136] 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 a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, an emotion recognition means that captures the user's facial expressions and analyzes their emotions, and an emotion input means that inputs the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model. This enables a music experience based on the user's real-time emotions and activities.

[1137] A "music providing means" is a means that includes a generative artificial intelligence model that generates a music playlist based on user preferences.

[1138] The "interaction means" is a means for updating the music playlist in real time by interacting with the user.

[1139] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[1140] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[1141] An "emotion recognition means" is a means for capturing a user's facial expressions and analyzing their emotions.

[1142] The "emotion input means" is a means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[1143] A "generative AI model" is an AI model equipped with an algorithm that generates playlists based on user preferences and updates them based on real-time data.

[1144] A "music playlist" is a list of music generated based on a user's preferences and emotions.

[1145] This invention provides a system that generates a music playlist based on a user's preferences and updates the playlist in real time. The system uses emotional data obtained from the user's current mood, activity status, and facial expressions to provide appropriate music content in a personalized manner. A specific embodiment of this invention will be described below.

[1146] Hardware and software used

[1147] Hardware:

[1148] Webcam: To capture the user's facial expressions

[1149] Smartphones and tablets: To interact with the user and play music

[1150] software:

[1151] OpenCV: A library for capturing and analyzing facial expressions

[1152] TensorFlow: A library for running emotion recognition models

[1153] requests: An HTTP request library for sending data to servers and retrieving music playlists.

[1154] System configuration

[1155] 1. Generative AI model that generates music playlists based on user preferences: Equipped with an AI model that generates optimal playlists based on the music a user has played in the past and their profile information.

[1156] 2. Interaction: A means to interact with the user in real time and update the music playlist. The user makes requests via text or voice input.

[1157] 3. Profile collection means: A means of obtaining a user's current mood and activity status, including information the user enters into the device and daily activity information.

[1158] 4. Input means: A means for inputting data collected by the profile collection means into the generative artificial intelligence model.

[1159] 5. Emotion Recognition: A means of capturing the user's facial expressions and analyzing their emotions. Specifically, facial images taken with a webcam are preprocessed using OpenCV and then analyzed using TensorFlow's emotion recognition model.

[1160] 6. Emotion input means: A means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[1161] Process Overview

[1162] The server first generates a music playlist based on the user's preferences. This process is based on past music playback history, current mood, activity status, and emotional data. The webcam captures the user's facial expressions and analyzes their emotions. This analysis, along with their profile information, is then fed into a generative artificial intelligence model to provide the optimal music playlist.

[1163] Examples of concrete examples and prompts

[1164] Examples:

[1165] The scenario imagines a user wearing smart glasses and a camera capturing their facial expressions.

[1166] The emotion recognition model detects when the user is "happy."

[1167] The server then sends the user a playlist of "relaxing music that matches a happy mood" along with the emotional data.

[1168] Example prompt sentence:

[1169] User ID: user12345, current emotion: happy

[1170] Recommend the best music content for this user.

[1171] This system enables a music experience based on the user's real-time emotions and activities, providing a deeper, more personalized entertainment experience.

[1172] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1173] Step 1:

[1174] A user launches a music streaming application and logs in. The user opens the app, enters their login information, and accesses their account. During this process, a set of initial data is collected, allowing the system to obtain the user's identity and profile information.

[1175] Step 2:

[1176] The device asks the user about their current mood and activity. Specifically, the app displays a pop-up asking questions such as "How are you feeling right now?" and "What are you currently doing?" to prompt the user for input. In response, the user provides information such as "I want to relax" or "I'm exercising." This information is collected by the profile collection means.

[1177] Step 3:

[1178] The device uses a webcam to capture the user's facial expressions. The emotion recognition unit preprocesses this facial image and converts it to grayscale. It then uses a face detection model to identify the face's location. The identified facial features are input into TensorFlow's emotion recognition model to analyze the user's emotions. The analysis results in emotional data such as "happiness," "anger," or "surprise."

[1179] Step 4:

[1180] The device sends the collected mood and activity data, as well as the analyzed emotional data, to the server, which then structures the data along with the profile information and sends it to the server, which receives it and prepares the data for real-time analysis.

[1181] Step 5:

[1182] The server inputs the received data into a generative AI model. The model is fed with the user's current mood, activity, emotions, and past music playback history. The generative AI model generates an optimal music playlist based on this data. The model analyzes the data and selects music to generate an optimal song list.

[1183] Step 6:

[1184] The server sends the generated music playlist to the device. The playlist is structured in JSON format and sent to the user's device. The device receives the playlist and displays it to the user.

[1185] Step 7:

[1186] The user plays songs from the generated playlist, and the device plays the selected songs and collects user feedback during the process. Feedback information is collected through simple form-filling and sentiment analysis.

[1187] Step 8:

[1188] The device sends the collected feedback and real-time emotional data to the server, which then inputs it back into the generative AI model to regenerate or update the playlist. The updated playlist is then sent back to the device and displayed to the user. This cycle creates a dynamic music experience for the user.

[1189] 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.

[1190] 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.

[1191] 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.

[1192] [Third embodiment]

[1193] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1194] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1195] 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).

[1196] 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.

[1197] 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.

[1198] 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).

[1199] 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.

[1200] 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.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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."

[1205] Collecting user profiles

[1206] 1. A user starts the music streaming application and begins using the system by logging in.

[1207] Example: A user opens an app and enters their login details to access their account.

[1208] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[1209] Example: An app pops up a question to the user, such as "How are you feeling right now?" or "What are your current activities?"

[1210] 3. The user enters the information into the terminal.

[1211] Example: A user enters information such as "I want to relax" or "I'm exercising."

[1212] 4. The device retrieves past music playback history from local storage and sends it to the server.

[1213] Example: An app gets a list of previously played songs and sends it to a server.

[1214] Data analysis and playlist generation

[1215] 1. The server receives the user's preferences and current state data.

[1216] Example: Receiving data such as "User ID: 12345, current mood: relaxed, activity: exercising."

[1217] 2. The server loads the generative artificial intelligence model and inputs the user profile.

[1218] Example: Providing profile information such as "relaxed" or "exercising" to an AI model.

[1219] 3. The server uses this data to generate a music playlist that best suits the user's current situation.

[1220] Example: Generate a playlist of relaxing and exercise-friendly music such as "Lo-fi Hip Hop" and "House Music."

[1221] 4. The server sends the generated playlist to the user's device.

[1222] Example: The generated playlist is sent to the app and displayed to the user.

[1223] Real-time playlist updates

[1224] 1. Listen to songs from a user-provided playlist.

[1225] Example: A user starts playing a song from a recommended playlist.

[1226] 2. The user inputs feedback about the song being played into the device, such as whether they like it or not, or if they want to skip it.

[1227] Example: Enter feedback such as "I like this song" or "I want to hear the next song."

[1228] 3. The device sends the feedback to the server.

[1229] Example: An app sends feedback information to a server in real time.

[1230] 4. The server regenerates or updates the playlist based on the feedback.

[1231] Example: Adding or changing songs in a playlist based on new feedback.

[1232] 5. The server sends the updated playlist back to the user's device.

[1233] Example: The updated playlist is reflected and displayed on the user's device.

[1234] User interaction

[1235] 1. The user makes a request or asks the AI ​​DJ a question, such as "What song do you recommend next?"

[1236] Example: A user types the question "What song should I play next?"

[1237] 2. The device sends the user's request or question to the server.

[1238] Example: An app sends a message from the user to a server.

[1239] 3. The server uses the AI ​​DJ model to generate an appropriate response.

[1240] Example: An AI DJ generates messages such as, "The song I recommend for your next relaxation session is XX."

[1241] 4. The server generates a response and sends it to the user's device.

[1242] Example: A response message is displayed in the user's app.

[1243] 5. The device displays or plays the AI ​​DJ's response to the user.

[1244] For example: The generated response is displayed or played in text or audio format.

[1245] This system allows users to get the best possible music experience based on their preferences and current situation, and allows for a more personalized music experience through real-time feedback and interaction.

[1246] The processing flow will be explained below.

[1247] Program processing steps

[1248] ---

[1249] Collecting user profiles

[1250] Step 1:

[1251] A user launches a music streaming application and logs in.

[1252] Specific behavior: Enter your username and password on the app's login screen and submit your authentication information.

[1253] Step 2:

[1254] The device asks the user about their current mood and activity.

[1255] Specific behavior: The app prompts the user with questions such as "How are you feeling right now?" and "What are you currently doing?"

[1256] Step 3:

[1257] The user inputs their current mood and activity status into the terminal.

[1258] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[1259] Step 4:

[1260] The device retrieves past music playback history from local storage and sends it to the server.

[1261] Specific operation: Reads playback history data from local storage and sends it to the server along with collected mood and activity information.

[1262] ---

[1263] Data analysis and playlist generation

[1264] Step 5:

[1265] The server receives the user's current mood, activity, and playback history data.

[1266] Specific operation: The server saves the received data in the database and begins analysis.

[1267] Step 6:

[1268] The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[1269] Specific operation: Create an instance of the AI ​​model and pass profile data such as "relaxed" or "exercising" as input parameters.

[1270] Step 7:

[1271] The server uses an AI model to generate a playlist based on the user's profile.

[1272] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile.

[1273] Step 8:

[1274] The server transmits the generated playlist to the user's terminal.

[1275] Specific operation: The selected song list is structured in JSON format and sent to the user's device.

[1276] ---

[1277] Real-time playlist updates

[1278] Step 9:

[1279] The terminal displays the generated playlist to the user.

[1280] Specific behavior: Parse the transmitted playlist data and display it in the app interface.

[1281] Step 10:

[1282] The user plays a song from the playlist.

[1283] Specific behavior: The song selected by the user will be played within the app.

[1284] Step 11:

[1285] The user inputs feedback about the song into the terminal.

[1286] Specific action: Enter feedback such as "I like this song" or "I want to hear the next song" into the input form.

[1287] Step 12:

[1288] The terminal transmits the feedback from the user to the server.

[1289] Specific behavior: The input feedback is sent to the server in real time.

[1290] Step 13:

[1291] The server regenerates or updates the playlist based on the feedback.

[1292] Specific behavior: Analyzes the received feedback data and performs processing to update the playlist.

[1293] Step 14:

[1294] The server sends the updated playlist to the user's terminal.

[1295] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[1296] ---

[1297] User interaction

[1298] Step 15:

[1299] Users make requests and ask questions to the AI ​​DJ.

[1300] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[1301] Step 16:

[1302] The terminal transmits the user's requests and questions to the server.

[1303] Specific operation: Sends the entered message data to the server.

[1304] Step 17:

[1305] The server uses an AI DJ model to generate appropriate responses.

[1306] Specific operation: The AI ​​DJ model analyzes the request content and generates an appropriate response message.

[1307] Step 18:

[1308] The server generates a response and sends it to the user's terminal.

[1309] Specific operation: The generated response message is sent to the user's terminal.

[1310] Step 19:

[1311] The terminal displays or plays the AI ​​DJ's responses to the user.

[1312] Specific behavior: The response message sent is displayed on the screen or played aloud.

[1313] Example 1

[1314] 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."

[1315] Conventional music streaming systems have struggled to provide real-time music playlists that respond to users' preferences, current moods, and activity status. Furthermore, they lack the ability to respond immediately to user feedback and requests, making it impossible to provide an optimized music experience for each individual user. The present invention aims to solve these problems and provide users with a more personalized music experience.

[1316] 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.

[1317] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model. This makes it possible to generate and update an optimal music playlist in real time based on the user's preferences and current status and provide it to the user.

[1318] "User" refers to an individual who uses the music streaming system.

[1319] "Preferences" refers to a user's musical tastes and tendencies.

[1320] "Mood" refers to the emotion or mental state a user is feeling at a particular time.

[1321] An "activity" is an action or task that a user is performing at a particular time.

[1322] "Music Playlist" means a list of songs organized for sequential playback by a user.

[1323] A "generative artificial intelligence model" refers to an artificial intelligence technology that generates new information or recommendations based on input data.

[1324] "Music provision means" refers to a mechanism that uses a generative artificial intelligence model to provide users with music playlists.

[1325] "Interaction means" refers to a mechanism by which users and systems communicate with each other.

[1326] "Profile Collection Measures" refers to mechanisms that collect information about a user's preferences, moods, and activities.

[1327] "Input means" refers to a mechanism for inputting data collected by the profile collection means into the generative artificial intelligence model.

[1328] "Server" refers to the computer system that processes user data and runs the generative artificial intelligence model.

[1329] This invention relates to a system that generates and updates an optimal music playlist in real time according to a user's preferences, current mood, and activity status. Specific embodiments of this system are described in detail below.

[1330] The system is mainly composed of three entities: a server, a terminal, and a user.

[1331] Collecting user profiles

[1332] A user begins using the system by launching a music streaming application and logging in. The user opens the app on a device such as a smartphone or tablet and enters their login information for authentication. If authentication is successful, the device displays a pop-up window asking about the user's current mood and activity. When the user enters information such as "I want to relax" or "I'm exercising," the device collects this information and retrieves past music playback history from local storage.

[1333] Data analysis and playlist generation

[1334] The data collected by the device is sent to the server. The server loads a generative artificial intelligence model and inputs the user profile based on the received user preferences and current state data. The generative AI model uses this data to generate a music playlist that is optimal for the user's current situation. The generated playlist is then sent back to the user's device, where it is displayed.

[1335] Real-time playlist updates

[1336] While listening to songs in a playlist, a user can input feedback about the song being played into the device. If the user inputs feedback such as "I like this song" or "I want to hear the next song," the device sends this information to the server in real time. The server regenerates or updates the playlist based on this feedback and sends the new playlist to the user's device.

[1337] User interaction

[1338] Users can make requests or questions to the AI ​​DJ. For example, if they input a question like "What's the next song you recommend?", the device will send this request to the server. The server will use the AI ​​DJ model to generate an appropriate response and send it to the user's device. The device will then display or play this response in text or audio format.

[1339] The system allows users to get the best possible music experience in real time based on their preferences and current situation, and further personalizes the music experience through user feedback and interaction.

[1340] Examples of prompt statements

[1341] Prompt: "What are the steps for a system that allows a user to enter their current mood and activity in a music streaming app and have AI generate the perfect playlist?"

[1342] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1343] Step 1:

[1344] The user launches the app and logs in

[1345] A user launches a music streaming app on a device such as a smartphone or tablet and enters their login information (email address, password). The device sends this input data to the server, which then authenticates the user. If authentication is successful, the server returns an authentication success message to the device, allowing the user to access their account.

[1346] Input: Email address, password

[1347] Output: Authentication success message

[1348] Step 2:

[1349] The device asks about the user's mood and activity

[1350] After logging in, the device displays a pop-up window asking the user questions such as "How are you feeling now?" and "What is your current activity?" To make it easier for the user to answer, options (e.g., relaxing, exercising, etc.) are displayed. These options may also be provided by the server.

[1351] Input: Authentication success message

[1352] Output: Mood and activity question popup

[1353] Step 3:

[1354] The user enters information

[1355] The user selects and inputs their mood and activity (e.g., "I want to relax" or "I'm exercising") from a pop-up window on the device, and the device immediately sends this information to the server.

[1356] Input: Questions about mood and activity

[1357] Output: User's mood and activity information

[1358] Step 4:

[1359] The device acquires past music playback history and sends it to the server.

[1360] The device searches for and retrieves past music playback history from its internal local storage, and the retrieved data is sent to the server.

[1361] Input: User's mood and activity information

[1362] Output: Past music playback history

[1363] Step 5:

[1364] The server receives the user's preferences and current state data.

[1365] The server receives data packets from the device about the user's mood, activity, and playback history, stores them in a database, and prepares them for analysis.

[1366] Input: User's mood and activity information, past music playback history

[1367] Output: Data stored in the database

[1368] Step 6:

[1369] The server loads the generative artificial intelligence model and inputs the user profile.

[1370] The server loads the generative AI model file from disk and expands it into memory. Data on the user's current mood, activity, and playback history is input into the generative AI model, and analysis begins.

[1371] Input: Data stored in a database

[1372] Output: User profile fed into the AI ​​model

[1373] Step 7:

[1374] The server generates a music playlist that best suits the user's current situation.

[1375] The server uses an AI model to select the best songs based on the user's input data and generate a playlist, which is then packaged into packets and sent to the device.

[1376] Input: User profile fed into the AI ​​model

[1377] Output: Generated music playlist

[1378] Step 8:

[1379] The server sends the generated playlist to the device.

[1380] The server assembles the generated playlist into a data packet and sends it to the user's terminal, which receives it and displays it to the user.

[1381] Input: Generated music playlist

[1382] Output: Playlist displayed on device

[1383] Step 9:

[1384] Listen to songs from a user-provided playlist

[1385] The user presses the play button on the playlist screen within the app to begin listening to the song, and the device responds by streaming the song data from the server.

[1386] Input: Playlist displayed on device

[1387] Output: The song being played

[1388] Step 10:

[1389] Users provide feedback on songs

[1390] Users can input feedback by tapping buttons on the playback screen such as "I like this song" or "I want to listen to the next song." The device then sends this feedback information to the server in real time.

[1391] Input: User feedback

[1392] Output: Feedback information sent to the server

[1393] Step 11:

[1394] The server regenerates or updates the playlist based on the feedback

[1395] The server re-runs the AI ​​model based on the received feedback, updates the playlist, and reconstructs the newly generated playlist and sends it to the device.

[1396] Input: Feedback information sent to the server

[1397] Output: Regenerated music playlist

[1398] Step 12:

[1399] The server sends the updated playlist to the device.

[1400] The server packages the updated playlist into a data packet and sends it to the user's terminal, which receives it and displays the new playlist.

[1401] Input: Regenerated music playlist

[1402] Output: Updated playlist displayed on device

[1403] Step 13:

[1404] Users make requests and ask questions to the AI ​​DJ

[1405] Users can type or speak using the in-app chat box or voice assistant function, asking, "What's your recommendation for the next song?"

[1406] Input: User requests or questions

[1407] Output: Requests and questions typed into the terminal

[1408] Step 14:

[1409] The device sends requests and questions to the server

[1410] The device sends the user's request or question as a data packet to the server, which receives the data and begins analyzing it.

[1411] Input: Requests or questions typed into the device

[1412] Output: The request or question sent to the server

[1413] Step 15:

[1414] The server generates a response using the AI ​​DJ model

[1415] The server runs the AI ​​DJ model to generate appropriate responses to requests and questions, which are then packaged into data packets and sent to the device.

[1416] Input: The request or question sent to the server

[1417] Output: The generated response message

[1418] Step 16:

[1419] Sends the server-generated response to the terminal

[1420] The server generates a response message and sends it as a data packet to the terminal, which receives it and displays or plays it audibly to the user.

[1421] Input: The generated response message

[1422] Output: The greeting message displayed or played on the terminal

[1423] (Application example 1)

[1424] 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."

[1425] Conventional food delivery systems have struggled to recommend the best food and restaurant in real time based on a user's preferences and current mood. Furthermore, they lacked the technology to instantly update recommendations based on user feedback. This resulted in a lack of consistency and individualization in the user experience.

[1426] 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.

[1427] In this invention, the server includes a recommendation means equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences, an interaction means that updates the recommendations in real time by interacting with the user, a profile collection means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collection means into the generative artificial intelligence model. This makes it possible to recommend optimal dishes and restaurants in real time based on the user's mood and activity status, and to instantly update the recommendations in response to feedback.

[1428] A "recommendation vehicle" is a vehicle equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences.

[1429] An "interaction means" is a means for updating recommendations in real time by interacting with the user.

[1430] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[1431] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[1432] A "generative artificial intelligence model" is an artificial intelligence model that generates optimal recommendations based on user preferences.

[1433] "Past order history" refers to the history of orders previously placed by the user.

[1434] An "algorithm" is a set of steps or rules for solving a specific problem by following a set of steps.

[1435] The system for implementing this invention mainly comprises a user terminal, a server, and a generative artificial intelligence model. The user terminal includes a mobile device such as a smartphone, and the server is located in a remote data center. The system has the function of collecting user profile information and providing optimal recommendations based on that information.

[1436] First, when a user logs in to the application using their smartphone, the device asks about the user's current mood and activity status. When the user enters information such as "I want to relax" or "I'm at a party," the device retrieves this information from local storage and sends it to the server.

[1437] The server receives the user's current status and past order history. Specifically, data such as "User ID: 12345, current mood: relaxed, activity: partying" is received and input into a generative AI model. This generative AI model generates optimal recommendations based on the user profile.

[1438] The generative AI model then uses the user's profile information to generate a list of recommended dishes and restaurants. For example, if a user wants to relax, snacks and desserts might be recommended, while if they're at a party, pizza and alcoholic drinks might be recommended.

[1439] Once the recommendation list is generated, the server sends the information to the user's device and displays it on the device. The user selects from the recommendation list and enters feedback into the device. For example, the user may enter feedback such as "I like this dish" or "I'd like to try a different dish next time." The device then sends this feedback to the server in real time, and the server again uses the generative artificial intelligence model to update the recommendation list.

[1440] The hardware used includes smartphones (user devices) and servers (remote data centers), and the software includes Python, Flask, generative artificial intelligence model libraries (e.g., some_ai_lib), JSON, HTTP, etc.

[1441] For example, if the prompt information "Current opinion: Relax, Activity: Partying" is entered, the generative artificial intelligence model will provide a recommendation based on this prompt information, such as "The next recommended dish is sushi and a tapioca drink."

[1442] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1443] Step 1:

[1444] A user logs in to an application on a smartphone. The user enters login information and accesses the application. The device receives this login information and performs user authentication. The input is the login information, and the output is the authentication status (success or failure).

[1445] Step 2:

[1446] If authentication is successful, the device displays a pop-up asking the user about their current mood and activity status. The user inputs information such as "I want to relax" or "I'm at a party." The input is the user's mood and activity status, and the output is a data format containing this information.

[1447] Step 3:

[1448] The device retrieves the user's current mood and activity status from local storage and sends it to the server. The input is the data in local storage and newly entered information, and the output is the data format (e.g., JSON format) sent to the server.

[1449] Step 4:

[1450] The server receives the user's current mood, activity status, and past order history and inputs them into a generative artificial intelligence model. The inputs are the user's current mood, activity status, and past order history, and the output is the profile data that is fed into the model.

[1451] Step 5:

[1452] A generative artificial intelligence model uses a user's profile information to generate a list of recommendations for the most suitable dishes and restaurants. The input is the user profile information, and the output is a list of recommendations. Specifically, the AI ​​model processes the data by analyzing past history and current status, and then uses a matching algorithm to identify recommended items.

[1453] Step 6:

[1454] The server sends the generated recommendation list to the user terminal. The input is the generated recommendation list, and the output is the recommendation list displayed on the user terminal. The user terminal displays this list on its screen.

[1455] Step 7:

[1456] The user selects from the recommendation list and inputs feedback into the terminal, for example, "I like this dish" or "I'd like to try a different dish next time." The input is the user's feedback, and the output is the feedback data sent to the server.

[1457] Step 8:

[1458] The device sends real-time feedback to the server, which then uses the generative AI model to update the recommendation list. The input is the feedback data, and the output is an updated recommendation list. Specifically, the AI ​​model re-runs optimization based on the new feedback to calculate the optimal recommendations.

[1459] Step 9:

[1460] The server sends the updated recommendation list back to the user's device, and the user confirms the new recommendations. The input is the updated recommendation list, and the output is the content displayed on the device. Based on this information, the user can choose their next action.

[1461] Through these processing steps, real-time recommendations based on the user's mood and activity status are provided.

[1462] 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.

[1463] User profile collection and emotion recognition

[1464] 1. A user starts the music streaming application and logs in to begin using the system.

[1465] Example: A user opens an app and enters their login details to access their account.

[1466] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[1467] Example: The app prompts you with questions like, "How are you feeling right now?" or "What are you doing right now?"

[1468] 3. The user enters the information into the terminal, and the terminal collects the information.

[1469] Example: A user enters information such as "I want to relax" or "I'm exercising."

[1470] 4. The emotion engine installed on the device recognizes and collects the user's emotions.

[1471] Example: The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotional data such as "The user looks happy" or "The user is feeling stressed."

[1472] 5. The device retrieves past music playback history from local storage and sends it to the server.

[1473] Example: An app retrieves a list of previously played songs and sends it to a server along with current mood, activity, and emotion data.

[1474] Data analysis and playlist generation

[1475] 1. The server receives the user's current mood, activity, emotion, and playback history data.

[1476] Example: Receive data such as "User ID: 12345, current mood: relaxed, activity: exercising, emotion: happy."

[1477] 2. The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[1478] Example: Create an instance of an AI model and provide profile information such as "relaxed," "exercising," or "happy" as input parameters.

[1479] 3. The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[1480] Example: An AI model selects appropriate songs based on a given profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[1481] 4. The server sends the generated playlist to the user's device.

[1482] Example: The generated playlist is structured in JSON format and sent to the user's device.

[1483] Real-time playlist updates

[1484] 1. The device displays the generated playlist to the user.

[1485] Example: Displaying a list of transmitted playlist data on the screen.

[1486] 2. The user plays a song from the playlist.

[1487] Example: A song selected by the user is played.

[1488] 3. The user enters their feedback on the song into the device, and the emotion engine continuously monitors the user's emotions.

[1489] Example: You input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze your facial expressions and collects emotional data.

[1490] 4. The device sends the user's feedback and emotion data to the server.

[1491] Example: Sending feedback information and emotion data to a server in real time.

[1492] 5. The server regenerates or updates the playlist based on the feedback and sentiment data.

[1493] Example: Analyzing the received data and reselecting the song that best suits the user's current state and emotions.

[1494] 6. The server sends the updated playlist to the user's device.

[1495] Example: New playlist data is structured in JSON format and sent to the user's device for display.

[1496] User interaction

[1497] 1. The user makes a request or asks a question to the AI ​​DJ.

[1498] Example: Enter a question into a form: "What's the next song you recommend?"

[1499] 2. The device sends the user's request or question to the server.

[1500] Example: Sending entered message data to the server.

[1501] 3. The server uses the AI ​​DJ model and emotional data to generate an appropriate response.

[1502] Example: An AI DJ analyzes the request content and emotional information and generates a response message such as, "The next song I recommend to relax you is XX."

[1503] 4. The server generates a response and sends it to the user's device.

[1504] Example: A response message is sent to the user's terminal and displayed.

[1505] 5. The device displays or plays the AI ​​DJ's response to the user.

[1506] Example: The generated response is displayed as text on the user's screen or played in audio format.

[1507] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[1508] The processing flow will be explained below.

[1509] Program processing steps

[1510] ---

[1511] User profile collection and emotion recognition

[1512] Step 1:

[1513] A user launches a music streaming application and logs in.

[1514] What happens: The user enters their username and password on the app's login screen and submits their authentication information.

[1515] Step 2:

[1516] The device asks the user about their current mood and activity.

[1517] What it does: The app prompts you with a pop-up asking questions like "How are you feeling right now?" and "What are you doing right now?"

[1518] Step 3:

[1519] The user inputs their current mood and activity status into the terminal.

[1520] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[1521] Step 4:

[1522] The emotion engine installed in the device recognizes the user's emotions.

[1523] Specific operation: The device's camera takes a picture of the user's face and analyzes their facial expressions. From the analyzed data, emotions such as "happiness" and "stress" are recognized.

[1524] Step 5:

[1525] The device retrieves the user's past music playback history from local storage and transmits it to the server.

[1526] What it does: Reads playback history data stored in local storage and sends it to the server along with current mood, activity, and recognized emotion data.

[1527] ---

[1528] Data analysis and playlist generation

[1529] Step 6:

[1530] The server receives the user's current mood, activity, emotion, and playback history data.

[1531] Specific operation: The server saves the received data in the database and begins analysis.

[1532] Step 7:

[1533] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[1534] What it does: Create an instance of an AI model and provide the user's profile data (e.g., "relaxed," "exercising," "happy") as input parameters.

[1535] Step 8:

[1536] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[1537] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile and emotional data.

[1538] Step 9:

[1539] The server transmits the generated playlist to the user's terminal.

[1540] Specific operation: The generated playlist is structured in JSON format and sent to the user's device.

[1541] ---

[1542] Real-time playlist updates

[1543] Step 10:

[1544] The terminal displays the generated playlist to the user.

[1545] Specific operation: The transmitted playlist data is displayed on the screen as a list.

[1546] Step 11:

[1547] The user plays the songs in the playlist.

[1548] Specific behavior: Starts playing the song selected by the user.

[1549] Step 12:

[1550] The user inputs feedback about the song being played into the terminal, and the emotion engine continues to recognize the user's emotions.

[1551] Specific operation: Input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze the user's facial expressions and collects emotional data.

[1552] Step 13:

[1553] The terminal transmits the user's feedback and emotion data to the server.

[1554] Specific operation: The input feedback information and recognized emotion data are sent to the server in real time.

[1555] Step 14:

[1556] The server regenerates or updates the playlist based on the feedback and emotional data.

[1557] Specific behavior: Analyzes the received data and reselects the song that best suits the user's current situation and emotions.

[1558] Step 15:

[1559] The server sends the updated playlist to the user's terminal.

[1560] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[1561] ---

[1562] User interaction

[1563] Step 16:

[1564] Users make requests and ask questions to the AI ​​DJ.

[1565] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[1566] Step 17:

[1567] The terminal transmits the user's requests and questions to the server.

[1568] Specific operation: Sends the input request and question data to the server.

[1569] Step 18:

[1570] The server uses the AI ​​DJ model and recognized emotion data to generate appropriate responses.

[1571] How it works: The AI ​​DJ model analyzes the request content and emotional data, and generates a response message such as, "The next recommended song is XX."

[1572] Step 19:

[1573] The server generates a response and sends it to the user's terminal.

[1574] Specific operation: Sends a response message to the terminal.

[1575] Step 20:

[1576] The terminal displays or plays the AI ​​DJ's responses to the user.

[1577] What happens: A response message is displayed on the screen as text or played aloud.

[1578] Example 2

[1579] 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."

[1580] While conventional music streaming systems provide playlists that take user preferences into account, they are unable to provide music that instantly adapts to the user's current mood, activity status, or even emotional changes. This makes it difficult to provide the optimal music experience for the user. Furthermore, the lack of a means to adjust the music experience through real-time feedback or dialogue also hinders the quality of the personalized music experience.

[1581] 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.

[1582] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity status, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood, activity status, and emotions, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, and an emotion recognizing means that continuously monitors and recognizes the user's emotions and reflects the acquired emotion data in the music playlist. This makes it possible to instantly respond to the user's preferences, current mood, activity status, and emotional changes and provide an optimal, personalized music experience.

[1583] "User" means any individual or organization that uses this system.

[1584] "Preferences" are information that indicates the user's past choices or usage trends.

[1585] "Mood" is information that indicates the user's current mental or emotional state.

[1586] "Activity status" is information that indicates the specific actions and situations that a user is currently engaged in.

[1587] A "generative artificial intelligence model" is an artificial intelligence algorithm that makes predictions and recommendations based on given data.

[1588] "Music provision means" refers to a function that provides music to users using a generative artificial intelligence model.

[1589] "Interaction means" is a function for communicating with users in real time and updating music playlists.

[1590] The "profile collection means" is a means for acquiring the user's current mood, activity status, and emotions.

[1591] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[1592] "Emotion recognition means" refers to a means for monitoring and recognizing a user's emotions and generating emotion data.

[1593] "Data" refers to various information, including information about a user's mood, activities, preferences, emotions, etc.

[1594] "Feedback" refers to opinions and evaluations provided by users to the system.

[1595] "Real-time" refers to actions or processes that respond immediately and without delay.

[1596] The present invention relates to a system for generating music playlists based on a user's preferences and current mood and activity status, which recognizes and reflects the user's current emotions in the music playlist, and provides a personalized music experience through real-time feedback and interaction.

[1597] User profile collection and emotion recognition

[1598] The user begins using the system by launching a music streaming application and logging in. The device asks the user about their current mood and activity (e.g., studying, exercising, or just wanting to relax). The user enters this information into the device, which then collects it. An emotion engine installed on the device recognizes and collects the user's emotions. The emotion engine uses tools such as Affectiva or Microsoft Emotion API. The device retrieves past music playback history from local storage and sends it to the server.

[1599] Data analysis and playlist generation

[1600] The server receives the user's current mood, activity, emotion, and playback history data. The server loads a generative artificial intelligence model and inputs the user's profile data into the model. This model uses GPT-3 or BERT. The server uses this data to generate a music playlist that best suits the user's current situation and emotion. The generated playlist is structured in JSON format and sent to the user's device.

[1601] Real-time playlist updates

[1602] The device displays the generated playlist to the user. As the user plays songs from the playlist, they can enter their feedback about the songs into the device. The emotion engine continues to monitor the user's emotions and collect emotion data. The device sends the user's feedback and emotion data to the server. The server regenerates or updates the playlist based on the feedback and emotion data. The updated playlist is again sent to the user's device in JSON format and displayed.

[1603] User interaction

[1604] Interaction between the user and the system begins when the user makes a request or question to the AI ​​DJ. For example, the user enters a question into the input form, such as "What song do you recommend next?" The device sends this request or question to the server. The server uses the AI ​​DJ model and emotional data to generate an appropriate response and sends it to the user's device. The device then displays or plays the AI ​​DJ's response to the user.

[1605] Specific examples

[1606] The user opens the app and enters their login details to access their account.

[1607] The app prompts you to enter information by popping up questions such as "How are you feeling right now?" and "What are you doing now?"

[1608] The user inputs information such as "I want to relax" or "I'm exercising."

[1609] The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotion data.

[1610] An instance of the AI ​​model is generated and profile information such as "relaxed," "exercising," and "happy" is provided as input parameters.

[1611] The AI ​​model selects appropriate songs based on the specified profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[1612] The generated playlist is structured in JSON format and sent to the user's device.

[1613] The user types a question into the input form: "What song do you recommend next?"

[1614] The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next song we recommend to help you relax is XX."

[1615] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[1616] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1617] Step 1:

[1618] A user launches a music streaming application and logs in.

[1619] Specific operation: The user taps the app to launch it, and the login screen appears. The user enters their username and password and presses the "Login" button.

[1620] Input: Username, Password

[1621] Output: Login session established, user information obtained

[1622] Step 2:

[1623] The device displays a prompt asking the user about their current mood and activity.

[1624] What it does: The app displays a pop-up window, offering the user options such as "How are you feeling right now?" and "What are you currently doing?", as well as a text input field.

[1625] Input: None

[1626] Output: Questions about mood and activity

[1627] Step 3:

[1628] The user inputs their current mood and activity into the device, and the device collects that information.

[1629] Specific operation: The user enters information such as "I want to relax" or "I'm exercising" and presses the "Send" button. The device saves this information in local storage.

[1630] Input: User's mood, activity information

[1631] Output: Mood and activity information stored in local storage

[1632] Step 4:

[1633] The emotion engine installed in the device recognizes and collects the user's emotions.

[1634] How it works: The device's camera captures the user's face, and the facial expression analysis algorithm generates emotional data such as happiness, sadness, and stress, which is then stored in local storage.

[1635] Input: User's face image

[1636] Output: Parsed emotion data

[1637] Step 5:

[1638] The device retrieves past music playback history from local storage and sends it to the server.

[1639] What it does: The device reads past playback history from local storage and sends it to the server as a JSON-formatted data packet along with current mood, activity, and emotion data.

[1640] Input: Playback history, current mood, activity, and emotion data in local storage

[1641] Output: Data packet sent to the server

[1642] Step 6:

[1643] The server receives the user's current mood, activity, emotion, and playback history data.

[1644] Specific operation: The server stores data packets received via the API endpoint in a database.

[1645] Input: User mood, activity, emotion, and playback history data

[1646] Output: User data stored in the database

[1647] Step 7:

[1648] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[1649] What it does: The server loads an instance of a generative AI model (e.g., GPT-3) into memory and provides the profile data as input parameters.

[1650] Input: User profile data

[1651] Output: Profile data input to the model

[1652] Step 8:

[1653] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[1654] What it does: The model analyzes the input data and lists the best songs. The playlist is structured in JSON format.

[1655] Input: Profile data input to the model

[1656] Output: Structured playlist data

[1657] Step 9:

[1658] The server transmits the generated playlist to the user's terminal.

[1659] Specific operation: Playlist data is sent to the user's device via API, and the device saves the received data.

[1660] Input: Structured playlist data

[1661] Output: Playlist data sent to the user's device

[1662] Step 10:

[1663] The terminal displays the generated playlist to the user.

[1664] Specific operation: Based on the received playlist data, the app displays a song list on the screen.

[1665] Input: Playlist data received from the server

[1666] Output: A list of songs displayed on the screen

[1667] Step 11:

[1668] The user plays a song from the playlist.

[1669] Specific behavior: When the user selects a song and presses the "Play" button, the music will play.

[1670] Input: User song selection

[1671] Output: Played music

[1672] Step 12:

[1673] The user inputs feedback about the song into the terminal, and the emotion engine continuously monitors the user's emotions.

[1674] Specific operation: The user inputs feedback such as "I like this song" or "I want to hear the next song." The device's camera continuously analyzes the user's facial expressions and generates emotion data.

[1675] Input: User feedback, face image

[1676] Output: Feedback data, analyzed emotion data

[1677] Step 13:

[1678] The device transmits the user's feedback and emotional data to the server.

[1679] What it does: Sends feedback and emotion data in real time to a server via an API.

[1680] Input: Feedback data, parsed emotion data

[1681] Output: Feedback and emotion data sent to the server

[1682] Step 14:

[1683] The server regenerates or updates the playlist based on the feedback and emotional data.

[1684] What happens: The server re-runs the AI ​​model based on the new data and generates an updated playlist.

[1685] Input: Feedback and emotion data

[1686] Output: Regenerated or updated playlist data

[1687] Step 15:

[1688] The server sends the updated playlist to the user's terminal.

[1689] Specific behavior: The new playlist data is sent in JSON format to the user's device, and the device displays the playlist again.

[1690] Input: Updated playlist data

[1691] Output: Updated playlist data sent to the user device

[1692] Step 16:

[1693] Users make requests and ask questions to the AI ​​DJ.

[1694] What happens: A user types a question into a text input form, such as "What's the next song you recommend?"

[1695] Input: User request or question

[1696] Output: The input request or question

[1697] Step 17:

[1698] The terminal transmits the user's requests and questions to the server.

[1699] Specific operation: The entered message data is sent to the server via the API.

[1700] Input: Request or question data

[1701] Output: Message data sent to the server

[1702] Step 18:

[1703] The server uses the AI ​​DJ model and emotional data to generate appropriate responses.

[1704] Specific operation: The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next recommended song to help you relax is XX."

[1705] Input: Request or question data, sentiment data

[1706] Output: The generated response message

[1707] Step 19:

[1708] The server generates a response and sends it to the user's terminal.

[1709] Specific operation: A response message is sent to the user's device via the API.

[1710] Input: The generated response message

[1711] Output: Response message sent to the user terminal

[1712] Step 20:

[1713] The terminal displays or plays the AI ​​DJ's responses to the user.

[1714] Specific Behavior: The generated response is displayed to the user in text on their screen or played in audio format.

[1715] Input: Response message sent to the user terminal

[1716] Output: The response message displayed or played on the user's terminal.

[1717] This allows users to get the best possible music experience based on their preferences and emotional information, and allows them to enjoy a more personalized music experience through real-time feedback and interaction.

[1718] (Application example 2)

[1719] 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."

[1720] Conventional music streaming services generate playlists based on users' music preferences, but they do not adequately provide a personalized music experience based on users' real-time emotions and activities. It is also difficult to instantly update playlists to reflect user feedback and emotional data in real time. As a result, it is difficult to provide users with the optimal music experience.

[1721] 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 a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, an emotion recognition means that captures the user's facial expressions and analyzes their emotions, and an emotion input means that inputs the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model. This enables a music experience based on the user's real-time emotions and activities.

[1722] A "music providing means" is a means that includes a generative artificial intelligence model that generates a music playlist based on user preferences.

[1723] The "interaction means" is a means for updating the music playlist in real time by interacting with the user.

[1724] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[1725] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[1726] An "emotion recognition means" is a means for capturing a user's facial expressions and analyzing their emotions.

[1727] The "emotion input means" is a means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[1728] A "generative AI model" is an AI model equipped with an algorithm that generates playlists based on user preferences and updates them based on real-time data.

[1729] A "music playlist" is a list of music generated based on a user's preferences and emotions.

[1730] This invention provides a system that generates a music playlist based on a user's preferences and updates the playlist in real time. The system uses emotional data obtained from the user's current mood, activity status, and facial expressions to provide appropriate music content in a personalized manner. A specific embodiment of this invention will be described below.

[1731] Hardware and software used

[1732] Hardware:

[1733] Webcam: To capture the user's facial expressions

[1734] Smartphones and tablets: To interact with the user and play music

[1735] software:

[1736] OpenCV: A library for capturing and analyzing facial expressions

[1737] TensorFlow: A library for running emotion recognition models

[1738] requests: An HTTP request library for sending data to servers and retrieving music playlists.

[1739] System configuration

[1740] 1. Generative AI model that generates music playlists based on user preferences: Equipped with an AI model that generates optimal playlists based on the music a user has played in the past and their profile information.

[1741] 2. Interaction: A means to interact with the user in real time and update the music playlist. The user makes requests via text or voice input.

[1742] 3. Profile collection means: A means of obtaining a user's current mood and activity status, including information the user enters into the device and daily activity information.

[1743] 4. Input means: A means for inputting data collected by the profile collection means into the generative artificial intelligence model.

[1744] 5. Emotion Recognition: A means of capturing the user's facial expressions and analyzing their emotions. Specifically, facial images taken with a webcam are preprocessed using OpenCV and then analyzed using TensorFlow's emotion recognition model.

[1745] 6. Emotion input means: A means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[1746] Process Overview

[1747] The server first generates a music playlist based on the user's preferences. This process is based on past music playback history, current mood, activity status, and emotional data. The webcam captures the user's facial expressions and analyzes their emotions. This analysis, along with their profile information, is then fed into a generative artificial intelligence model to provide the optimal music playlist.

[1748] Examples of concrete examples and prompts

[1749] Examples:

[1750] The scenario imagines a user wearing smart glasses and a camera capturing their facial expressions.

[1751] The emotion recognition model detects when the user is "happy."

[1752] The server then sends the user a playlist of "relaxing music that matches a happy mood" along with the emotional data.

[1753] Example prompt sentence:

[1754] User ID: user12345, current emotion: happy

[1755] Recommend the best music content for this user.

[1756] This system enables a music experience based on the user's real-time emotions and activities, providing a deeper, more personalized entertainment experience.

[1757] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1758] Step 1:

[1759] A user launches a music streaming application and logs in. The user opens the app, enters their login information, and accesses their account. During this process, a set of initial data is collected, allowing the system to obtain the user's identity and profile information.

[1760] Step 2:

[1761] The device asks the user about their current mood and activity. Specifically, the app displays a pop-up asking questions such as "How are you feeling right now?" and "What are you currently doing?" to prompt the user for input. In response, the user provides information such as "I want to relax" or "I'm exercising." This information is collected by the profile collection means.

[1762] Step 3:

[1763] The device uses a webcam to capture the user's facial expressions. The emotion recognition unit preprocesses this facial image and converts it to grayscale. It then uses a face detection model to identify the face's location. The identified facial features are input into TensorFlow's emotion recognition model to analyze the user's emotions. The analysis results in emotional data such as "happiness," "anger," or "surprise."

[1764] Step 4:

[1765] The device sends the collected mood and activity data, as well as the analyzed emotional data, to the server, which then structures the data along with the profile information and sends it to the server, which receives it and prepares the data for real-time analysis.

[1766] Step 5:

[1767] The server inputs the received data into a generative AI model. The model is fed with the user's current mood, activity, emotions, and past music playback history. The generative AI model generates an optimal music playlist based on this data. The model analyzes the data and selects music to generate an optimal song list.

[1768] Step 6:

[1769] The server sends the generated music playlist to the device. The playlist is structured in JSON format and sent to the user's device. The device receives the playlist and displays it to the user.

[1770] Step 7:

[1771] The user plays songs from the generated playlist, and the device plays the selected songs and collects user feedback during the process. Feedback information is collected through simple form-filling and sentiment analysis.

[1772] Step 8:

[1773] The device sends the collected feedback and real-time emotional data to the server, which then inputs it back into the generative AI model to regenerate or update the playlist. The updated playlist is then sent back to the device and displayed to the user. This cycle creates a dynamic music experience for the user.

[1774] 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.

[1775] 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.

[1776] 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.

[1777] [Fourth embodiment]

[1778] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1779] 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.

[1780] 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).

[1781] 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.

[1782] 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.

[1783] 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).

[1784] 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.

[1785] 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.

[1786] 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.

[1787] 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.

[1788] 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.

[1789] 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.

[1790] 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."

[1791] Collecting user profiles

[1792] 1. A user starts the music streaming application and begins using the system by logging in.

[1793] Example: A user opens an app and enters their login details to access their account.

[1794] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[1795] Example: An app pops up a question to the user, such as "How are you feeling right now?" or "What are your current activities?"

[1796] 3. The user enters the information into the terminal.

[1797] Example: A user enters information such as "I want to relax" or "I'm exercising."

[1798] 4. The device retrieves past music playback history from local storage and sends it to the server.

[1799] Example: An app gets a list of previously played songs and sends it to a server.

[1800] Data analysis and playlist generation

[1801] 1. The server receives the user's preferences and current state data.

[1802] Example: Receiving data such as "User ID: 12345, current mood: relaxed, activity: exercising."

[1803] 2. The server loads the generative artificial intelligence model and inputs the user profile.

[1804] Example: Providing profile information such as "relaxed" or "exercising" to an AI model.

[1805] 3. The server uses this data to generate a music playlist that best suits the user's current situation.

[1806] Example: Generate a playlist of relaxing and exercise-friendly music such as "Lo-fi Hip Hop" and "House Music."

[1807] 4. The server sends the generated playlist to the user's device.

[1808] Example: The generated playlist is sent to the app and displayed to the user.

[1809] Real-time playlist updates

[1810] 1. Listen to songs from a user-provided playlist.

[1811] Example: A user starts playing a song from a recommended playlist.

[1812] 2. The user inputs feedback about the song being played into the device, such as whether they like it or not, or if they want to skip it.

[1813] Example: Enter feedback such as "I like this song" or "I want to hear the next song."

[1814] 3. The device sends the feedback to the server.

[1815] Example: An app sends feedback information to a server in real time.

[1816] 4. The server regenerates or updates the playlist based on the feedback.

[1817] Example: Adding or changing songs in a playlist based on new feedback.

[1818] 5. The server sends the updated playlist back to the user's device.

[1819] Example: The updated playlist is reflected and displayed on the user's device.

[1820] User interaction

[1821] 1. The user makes a request or asks the AI ​​DJ a question, such as "What song do you recommend next?"

[1822] Example: A user types the question "What song should I play next?"

[1823] 2. The device sends the user's request or question to the server.

[1824] Example: An app sends a message from the user to a server.

[1825] 3. The server uses the AI ​​DJ model to generate an appropriate response.

[1826] Example: An AI DJ generates messages such as, "The song I recommend for your next relaxation session is XX."

[1827] 4. The server generates a response and sends it to the user's device.

[1828] Example: A response message is displayed in the user's app.

[1829] 5. The device displays or plays the AI ​​DJ's response to the user.

[1830] For example: The generated response is displayed or played in text or audio format.

[1831] This system allows users to get the best possible music experience based on their preferences and current situation, and allows for a more personalized music experience through real-time feedback and interaction.

[1832] The processing flow will be explained below.

[1833] Program processing steps

[1834] ---

[1835] Collecting user profiles

[1836] Step 1:

[1837] A user launches a music streaming application and logs in.

[1838] Specific behavior: Enter your username and password on the app's login screen and submit your authentication information.

[1839] Step 2:

[1840] The device asks the user about their current mood and activity.

[1841] Specific behavior: The app prompts the user with questions such as "How are you feeling right now?" and "What are you currently doing?"

[1842] Step 3:

[1843] The user inputs their current mood and activity status into the terminal.

[1844] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[1845] Step 4:

[1846] The device retrieves past music playback history from local storage and sends it to the server.

[1847] Specific operation: Reads playback history data from local storage and sends it to the server along with collected mood and activity information.

[1848] ---

[1849] Data analysis and playlist generation

[1850] Step 5:

[1851] The server receives the user's current mood, activity, and playback history data.

[1852] Specific operation: The server saves the received data in the database and begins analysis.

[1853] Step 6:

[1854] The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[1855] Specific operation: Create an instance of the AI ​​model and pass profile data such as "relaxed" or "exercising" as input parameters.

[1856] Step 7:

[1857] The server uses an AI model to generate a playlist based on the user's profile.

[1858] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile.

[1859] Step 8:

[1860] The server transmits the generated playlist to the user's terminal.

[1861] Specific operation: The selected song list is structured in JSON format and sent to the user's device.

[1862] ---

[1863] Real-time playlist updates

[1864] Step 9:

[1865] The terminal displays the generated playlist to the user.

[1866] Specific behavior: Parse the transmitted playlist data and display it in the app interface.

[1867] Step 10:

[1868] The user plays a song from the playlist.

[1869] Specific behavior: The song selected by the user will be played within the app.

[1870] Step 11:

[1871] The user inputs feedback about the song into the terminal.

[1872] Specific action: Enter feedback such as "I like this song" or "I want to hear the next song" into the input form.

[1873] Step 12:

[1874] The terminal transmits the feedback from the user to the server.

[1875] Specific behavior: The input feedback is sent to the server in real time.

[1876] Step 13:

[1877] The server regenerates or updates the playlist based on the feedback.

[1878] Specific behavior: Analyzes the received feedback data and performs processing to update the playlist.

[1879] Step 14:

[1880] The server sends the updated playlist to the user's terminal.

[1881] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[1882] ---

[1883] User interaction

[1884] Step 15:

[1885] Users make requests and ask questions to the AI ​​DJ.

[1886] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[1887] Step 16:

[1888] The terminal transmits the user's requests and questions to the server.

[1889] Specific operation: Sends the entered message data to the server.

[1890] Step 17:

[1891] The server uses an AI DJ model to generate appropriate responses.

[1892] Specific operation: The AI ​​DJ model analyzes the request content and generates an appropriate response message.

[1893] Step 18:

[1894] The server generates a response and sends it to the user's terminal.

[1895] Specific operation: The generated response message is sent to the user's terminal.

[1896] Step 19:

[1897] The terminal displays or plays the AI ​​DJ's responses to the user.

[1898] Specific behavior: The response message sent is displayed on the screen or played aloud.

[1899] Example 1

[1900] 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."

[1901] Conventional music streaming systems have struggled to provide real-time music playlists that respond to users' preferences, current moods, and activity status. Furthermore, they lack the ability to respond immediately to user feedback and requests, making it impossible to provide an optimized music experience for each individual user. The present invention aims to solve these problems and provide users with a more personalized music experience.

[1902] 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.

[1903] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model. This makes it possible to generate and update an optimal music playlist in real time based on the user's preferences and current status and provide it to the user.

[1904] "User" refers to an individual who uses the music streaming system.

[1905] "Preferences" refers to a user's musical tastes and tendencies.

[1906] "Mood" refers to the emotion or mental state a user is feeling at a particular time.

[1907] An "activity" is an action or task that a user is performing at a particular time.

[1908] "Music Playlist" means a list of songs organized for sequential playback by a user.

[1909] A "generative artificial intelligence model" refers to an artificial intelligence technology that generates new information or recommendations based on input data.

[1910] "Music provision means" refers to a mechanism that uses a generative artificial intelligence model to provide users with music playlists.

[1911] "Interaction means" refers to a mechanism by which users and systems communicate with each other.

[1912] "Profile Collection Measures" refers to mechanisms that collect information about a user's preferences, moods, and activities.

[1913] "Input means" refers to a mechanism for inputting data collected by the profile collection means into the generative artificial intelligence model.

[1914] "Server" refers to the computer system that processes user data and runs the generative artificial intelligence model.

[1915] This invention relates to a system that generates and updates an optimal music playlist in real time according to a user's preferences, current mood, and activity status. Specific embodiments of this system are described in detail below.

[1916] The system is mainly composed of three entities: a server, a terminal, and a user.

[1917] Collecting user profiles

[1918] A user begins using the system by launching a music streaming application and logging in. The user opens the app on a device such as a smartphone or tablet and enters their login information for authentication. If authentication is successful, the device displays a pop-up window asking about the user's current mood and activity. When the user enters information such as "I want to relax" or "I'm exercising," the device collects this information and retrieves past music playback history from local storage.

[1919] Data analysis and playlist generation

[1920] The data collected by the device is sent to the server. The server loads a generative artificial intelligence model and inputs the user profile based on the received user preferences and current state data. The generative AI model uses this data to generate a music playlist that is optimal for the user's current situation. The generated playlist is then sent back to the user's device, where it is displayed.

[1921] Real-time playlist updates

[1922] While listening to songs in a playlist, a user can input feedback about the song being played into the device. If the user inputs feedback such as "I like this song" or "I want to hear the next song," the device sends this information to the server in real time. The server regenerates or updates the playlist based on this feedback and sends the new playlist to the user's device.

[1923] User interaction

[1924] Users can make requests or questions to the AI ​​DJ. For example, if they input a question like "What's the next song you recommend?", the device will send this request to the server. The server will use the AI ​​DJ model to generate an appropriate response and send it to the user's device. The device will then display or play this response in text or audio format.

[1925] The system allows users to get the best possible music experience in real time based on their preferences and current situation, and further personalizes the music experience through user feedback and interaction.

[1926] Examples of prompt statements

[1927] Prompt: "What are the steps for a system that allows a user to enter their current mood and activity in a music streaming app and have AI generate the perfect playlist?"

[1928] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1929] Step 1:

[1930] The user launches the app and logs in

[1931] A user launches a music streaming app on a device such as a smartphone or tablet and enters their login information (email address, password). The device sends this input data to the server, which then authenticates the user. If authentication is successful, the server returns an authentication success message to the device, allowing the user to access their account.

[1932] Input: Email address, password

[1933] Output: Authentication success message

[1934] Step 2:

[1935] The device asks about the user's mood and activity

[1936] After logging in, the device displays a pop-up window asking the user questions such as "How are you feeling now?" and "What is your current activity?" To make it easier for the user to answer, options (e.g., relaxing, exercising, etc.) are displayed. These options may also be provided by the server.

[1937] Input: Authentication success message

[1938] Output: Mood and activity question popup

[1939] Step 3:

[1940] The user enters information

[1941] The user selects and inputs their mood and activity (e.g., "I want to relax" or "I'm exercising") from a pop-up window on the device, and the device immediately sends this information to the server.

[1942] Input: Questions about mood and activity

[1943] Output: User's mood and activity information

[1944] Step 4:

[1945] The device acquires past music playback history and sends it to the server.

[1946] The device searches for and retrieves past music playback history from its internal local storage, and the retrieved data is sent to the server.

[1947] Input: User's mood and activity information

[1948] Output: Past music playback history

[1949] Step 5:

[1950] The server receives the user's preferences and current state data.

[1951] The server receives data packets from the device about the user's mood, activity, and playback history, stores them in a database, and prepares them for analysis.

[1952] Input: User's mood and activity information, past music playback history

[1953] Output: Data stored in the database

[1954] Step 6:

[1955] The server loads the generative artificial intelligence model and inputs the user profile.

[1956] The server loads the generative AI model file from disk and expands it into memory. Data on the user's current mood, activity, and playback history is input into the generative AI model, and analysis begins.

[1957] Input: Data stored in a database

[1958] Output: User profile fed into the AI ​​model

[1959] Step 7:

[1960] The server generates a music playlist that best suits the user's current situation.

[1961] The server uses an AI model to select the best songs based on the user's input data and generate a playlist, which is then packaged into packets and sent to the device.

[1962] Input: User profile fed into the AI ​​model

[1963] Output: Generated music playlist

[1964] Step 8:

[1965] The server sends the generated playlist to the device.

[1966] The server assembles the generated playlist into a data packet and sends it to the user's terminal, which receives it and displays it to the user.

[1967] Input: Generated music playlist

[1968] Output: Playlist displayed on device

[1969] Step 9:

[1970] Listen to songs from a user-provided playlist

[1971] The user presses the play button on the playlist screen within the app to begin listening to the song, and the device responds by streaming the song data from the server.

[1972] Input: Playlist displayed on device

[1973] Output: The song being played

[1974] Step 10:

[1975] Users provide feedback on songs

[1976] Users can input feedback by tapping buttons on the playback screen such as "I like this song" or "I want to listen to the next song." The device then sends this feedback information to the server in real time.

[1977] Input: User feedback

[1978] Output: Feedback information sent to the server

[1979] Step 11:

[1980] The server regenerates or updates the playlist based on the feedback

[1981] The server re-runs the AI ​​model based on the received feedback, updates the playlist, and reconstructs the newly generated playlist and sends it to the device.

[1982] Input: Feedback information sent to the server

[1983] Output: Regenerated music playlist

[1984] Step 12:

[1985] The server sends the updated playlist to the device.

[1986] The server packages the updated playlist into a data packet and sends it to the user's terminal, which receives it and displays the new playlist.

[1987] Input: Regenerated music playlist

[1988] Output: Updated playlist displayed on device

[1989] Step 13:

[1990] Users make requests and ask questions to the AI ​​DJ

[1991] Users can type or speak using the in-app chat box or voice assistant function, asking, "What's your recommendation for the next song?"

[1992] Input: User requests or questions

[1993] Output: Requests and questions typed into the terminal

[1994] Step 14:

[1995] The device sends requests and questions to the server

[1996] The device sends the user's request or question as a data packet to the server, which receives the data and begins analyzing it.

[1997] Input: Requests or questions typed into the device

[1998] Output: The request or question sent to the server

[1999] Step 15:

[2000] The server generates a response using the AI ​​DJ model

[2001] The server runs the AI ​​DJ model to generate appropriate responses to requests and questions, which are then packaged into data packets and sent to the device.

[2002] Input: The request or question sent to the server

[2003] Output: The generated response message

[2004] Step 16:

[2005] Sends the server-generated response to the terminal

[2006] The server generates a response message and sends it as a data packet to the terminal, which receives it and displays or plays it audibly to the user.

[2007] Input: The generated response message

[2008] Output: The greeting message displayed or played on the terminal

[2009] (Application example 1)

[2010] 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."

[2011] Conventional food delivery systems have struggled to recommend the best food and restaurant in real time based on a user's preferences and current mood. Furthermore, they lacked the technology to instantly update recommendations based on user feedback. This resulted in a lack of consistency and individualization in the user experience.

[2012] 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.

[2013] In this invention, the server includes a recommendation means equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences, an interaction means that updates the recommendations in real time by interacting with the user, a profile collection means that acquires the user's current mood and activity status, and an input means that inputs the data collected by the profile collection means into the generative artificial intelligence model. This makes it possible to recommend optimal dishes and restaurants in real time based on the user's mood and activity status, and to instantly update the recommendations in response to feedback.

[2014] A "recommendation vehicle" is a vehicle equipped with a generative artificial intelligence model that generates optimal recommendations based on user preferences.

[2015] An "interaction means" is a means for updating recommendations in real time by interacting with the user.

[2016] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[2017] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[2018] A "generative artificial intelligence model" is an artificial intelligence model that generates optimal recommendations based on user preferences.

[2019] "Past order history" refers to the history of orders previously placed by the user.

[2020] An "algorithm" is a set of steps or rules for solving a specific problem by following a set of steps.

[2021] The system for implementing this invention mainly comprises a user terminal, a server, and a generative artificial intelligence model. The user terminal includes a mobile device such as a smartphone, and the server is located in a remote data center. The system has the function of collecting user profile information and providing optimal recommendations based on that information.

[2022] First, when a user logs in to the application using their smartphone, the device asks about the user's current mood and activity status. When the user enters information such as "I want to relax" or "I'm at a party," the device retrieves this information from local storage and sends it to the server.

[2023] The server receives the user's current status and past order history. Specifically, data such as "User ID: 12345, current mood: relaxed, activity: partying" is received and input into a generative AI model. This generative AI model generates optimal recommendations based on the user profile.

[2024] The generative AI model then uses the user's profile information to generate a list of recommended dishes and restaurants. For example, if a user wants to relax, snacks and desserts might be recommended, while if they're at a party, pizza and alcoholic drinks might be recommended.

[2025] Once the recommendation list is generated, the server sends the information to the user's device and displays it on the device. The user selects from the recommendation list and enters feedback into the device. For example, the user may enter feedback such as "I like this dish" or "I'd like to try a different dish next time." The device then sends this feedback to the server in real time, and the server again uses the generative artificial intelligence model to update the recommendation list.

[2026] The hardware used includes smartphones (user devices) and servers (remote data centers), and the software includes Python, Flask, generative artificial intelligence model libraries (e.g., some_ai_lib), JSON, HTTP, etc.

[2027] For example, if the prompt information "Current opinion: Relax, Activity: Partying" is entered, the generative artificial intelligence model will provide a recommendation based on this prompt information, such as "The next recommended dish is sushi and a tapioca drink."

[2028] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2029] Step 1:

[2030] A user logs in to an application on a smartphone. The user enters login information and accesses the application. The device receives this login information and performs user authentication. The input is the login information, and the output is the authentication status (success or failure).

[2031] Step 2:

[2032] If authentication is successful, the device displays a pop-up asking the user about their current mood and activity status. The user inputs information such as "I want to relax" or "I'm at a party." The input is the user's mood and activity status, and the output is a data format containing this information.

[2033] Step 3:

[2034] The device retrieves the user's current mood and activity status from local storage and sends it to the server. The input is the data in local storage and newly entered information, and the output is the data format (e.g., JSON format) sent to the server.

[2035] Step 4:

[2036] The server receives the user's current mood, activity status, and past order history and inputs them into a generative artificial intelligence model. The inputs are the user's current mood, activity status, and past order history, and the output is the profile data that is fed into the model.

[2037] Step 5:

[2038] A generative artificial intelligence model uses a user's profile information to generate a list of recommendations for the most suitable dishes and restaurants. The input is the user profile information, and the output is a list of recommendations. Specifically, the AI ​​model processes the data by analyzing past history and current status, and then uses a matching algorithm to identify recommended items.

[2039] Step 6:

[2040] The server sends the generated recommendation list to the user terminal. The input is the generated recommendation list, and the output is the recommendation list displayed on the user terminal. The user terminal displays this list on its screen.

[2041] Step 7:

[2042] The user selects from the recommendation list and inputs feedback into the terminal, for example, "I like this dish" or "I'd like to try a different dish next time." The input is the user's feedback, and the output is the feedback data sent to the server.

[2043] Step 8:

[2044] The device sends real-time feedback to the server, which then uses the generative AI model to update the recommendation list. The input is the feedback data, and the output is an updated recommendation list. Specifically, the AI ​​model re-runs optimization based on the new feedback to calculate the optimal recommendations.

[2045] Step 9:

[2046] The server sends the updated recommendation list back to the user's device, and the user confirms the new recommendations. The input is the updated recommendation list, and the output is the content displayed on the device. Based on this information, the user can choose their next action.

[2047] Through these processing steps, real-time recommendations based on the user's mood and activity status are provided.

[2048] 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.

[2049] User profile collection and emotion recognition

[2050] 1. A user starts the music streaming application and logs in to begin using the system.

[2051] Example: A user opens an app and enters their login details to access their account.

[2052] 2. The device asks the user about their current mood or activity (e.g., studying, exercising, wanting to relax, etc.).

[2053] Example: The app prompts you with questions like, "How are you feeling right now?" or "What are you doing right now?"

[2054] 3. The user enters the information into the terminal, and the terminal collects the information.

[2055] Example: A user enters information such as "I want to relax" or "I'm exercising."

[2056] 4. The emotion engine installed on the device recognizes and collects the user's emotions.

[2057] Example: The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotional data such as "The user looks happy" or "The user is feeling stressed."

[2058] 5. The device retrieves past music playback history from local storage and sends it to the server.

[2059] Example: An app retrieves a list of previously played songs and sends it to a server along with current mood, activity, and emotion data.

[2060] Data analysis and playlist generation

[2061] 1. The server receives the user's current mood, activity, emotion, and playback history data.

[2062] Example: Receive data such as "User ID: 12345, current mood: relaxed, activity: exercising, emotion: happy."

[2063] 2. The server loads the generative artificial intelligence model and inputs the user's profile data into the model.

[2064] Example: Create an instance of an AI model and provide profile information such as "relaxed," "exercising," or "happy" as input parameters.

[2065] 3. The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[2066] Example: An AI model selects appropriate songs based on a given profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[2067] 4. The server sends the generated playlist to the user's device.

[2068] Example: The generated playlist is structured in JSON format and sent to the user's device.

[2069] Real-time playlist updates

[2070] 1. The device displays the generated playlist to the user.

[2071] Example: Displaying a list of transmitted playlist data on the screen.

[2072] 2. The user plays a song from the playlist.

[2073] Example: A song selected by the user is played.

[2074] 3. The user enters their feedback on the song into the device, and the emotion engine continuously monitors the user's emotions.

[2075] Example: You input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze your facial expressions and collects emotional data.

[2076] 4. The device sends the user's feedback and emotion data to the server.

[2077] Example: Sending feedback information and emotion data to a server in real time.

[2078] 5. The server regenerates or updates the playlist based on the feedback and sentiment data.

[2079] Example: Analyzing the received data and reselecting the song that best suits the user's current state and emotions.

[2080] 6. The server sends the updated playlist to the user's device.

[2081] Example: New playlist data is structured in JSON format and sent to the user's device for display.

[2082] User interaction

[2083] 1. The user makes a request or asks a question to the AI ​​DJ.

[2084] Example: Enter a question into a form: "What's the next song you recommend?"

[2085] 2. The device sends the user's request or question to the server.

[2086] Example: Sending entered message data to the server.

[2087] 3. The server uses the AI ​​DJ model and emotional data to generate an appropriate response.

[2088] Example: An AI DJ analyzes the request content and emotional information and generates a response message such as, "The next song I recommend to relax you is XX."

[2089] 4. The server generates a response and sends it to the user's device.

[2090] Example: A response message is sent to the user's terminal and displayed.

[2091] 5. The device displays or plays the AI ​​DJ's response to the user.

[2092] Example: The generated response is displayed as text on the user's screen or played in audio format.

[2093] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[2094] The processing flow will be explained below.

[2095] Program processing steps

[2096] ---

[2097] User profile collection and emotion recognition

[2098] Step 1:

[2099] A user launches a music streaming application and logs in.

[2100] What happens: The user enters their username and password on the app's login screen and submits their authentication information.

[2101] Step 2:

[2102] The device asks the user about their current mood and activity.

[2103] What it does: The app prompts you with a pop-up asking questions like "How are you feeling right now?" and "What are you doing right now?"

[2104] Step 3:

[2105] The user inputs their current mood and activity status into the terminal.

[2106] Specific actions: Use a form or options to enter information such as "I want to relax" or "I'm exercising."

[2107] Step 4:

[2108] The emotion engine installed in the device recognizes the user's emotions.

[2109] Specific operation: The device's camera takes a picture of the user's face and analyzes their facial expressions. From the analyzed data, emotions such as "happiness" and "stress" are recognized.

[2110] Step 5:

[2111] The device retrieves the user's past music playback history from local storage and transmits it to the server.

[2112] What it does: Reads playback history data stored in local storage and sends it to the server along with current mood, activity, and recognized emotion data.

[2113] ---

[2114] Data analysis and playlist generation

[2115] Step 6:

[2116] The server receives the user's current mood, activity, emotion, and playback history data.

[2117] Specific operation: The server saves the received data in the database and begins analysis.

[2118] Step 7:

[2119] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[2120] What it does: Create an instance of an AI model and provide the user's profile data (e.g., "relaxed," "exercising," "happy") as input parameters.

[2121] Step 8:

[2122] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[2123] What it does: The AI ​​model selects appropriate songs and generates a playlist based on the specified profile and emotional data.

[2124] Step 9:

[2125] The server transmits the generated playlist to the user's terminal.

[2126] Specific operation: The generated playlist is structured in JSON format and sent to the user's device.

[2127] ---

[2128] Real-time playlist updates

[2129] Step 10:

[2130] The terminal displays the generated playlist to the user.

[2131] Specific operation: The transmitted playlist data is displayed on the screen as a list.

[2132] Step 11:

[2133] The user plays the songs in the playlist.

[2134] Specific behavior: Starts playing the song selected by the user.

[2135] Step 12:

[2136] The user inputs feedback about the song being played into the terminal, and the emotion engine continues to recognize the user's emotions.

[2137] Specific operation: Input feedback such as "I like this song" or "I want to hear the next song." The emotion engine continues to analyze the user's facial expressions and collects emotional data.

[2138] Step 13:

[2139] The terminal transmits the user's feedback and emotion data to the server.

[2140] Specific operation: The input feedback information and recognized emotion data are sent to the server in real time.

[2141] Step 14:

[2142] The server regenerates or updates the playlist based on the feedback and emotional data.

[2143] Specific behavior: Analyzes the received data and reselects the song that best suits the user's current situation and emotions.

[2144] Step 15:

[2145] The server sends the updated playlist to the user's terminal.

[2146] Specific behavior: Structure new playlist data in JSON format and resend it to the user's device.

[2147] ---

[2148] User interaction

[2149] Step 16:

[2150] Users make requests and ask questions to the AI ​​DJ.

[2151] What it does: Enter a question into a form, such as "What's the next song you recommend?"

[2152] Step 17:

[2153] The terminal transmits the user's requests and questions to the server.

[2154] Specific operation: Sends the input request and question data to the server.

[2155] Step 18:

[2156] The server uses the AI ​​DJ model and recognized emotion data to generate appropriate responses.

[2157] How it works: The AI ​​DJ model analyzes the request content and emotional data, and generates a response message such as, "The next recommended song is XX."

[2158] Step 19:

[2159] The server generates a response and sends it to the user's terminal.

[2160] Specific operation: Sends a response message to the terminal.

[2161] Step 20:

[2162] The terminal displays or plays the AI ​​DJ's responses to the user.

[2163] What happens: A response message is displayed on the screen as text or played aloud.

[2164] Example 2

[2165] 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."

[2166] While conventional music streaming systems provide playlists that take user preferences into account, they are unable to provide music that instantly adapts to the user's current mood, activity status, or even emotional changes. This makes it difficult to provide the optimal music experience for the user. Furthermore, the lack of a means to adjust the music experience through real-time feedback or dialogue also hinders the quality of the personalized music experience.

[2167] 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.

[2168] In this invention, the server includes a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences and current mood and activity status, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood, activity status, and emotions, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, and an emotion recognizing means that continuously monitors and recognizes the user's emotions and reflects the acquired emotion data in the music playlist. This makes it possible to instantly respond to the user's preferences, current mood, activity status, and emotional changes and provide an optimal, personalized music experience.

[2169] "User" means any individual or organization that uses this system.

[2170] "Preferences" are information that indicates the user's past choices or usage trends.

[2171] "Mood" is information that indicates the user's current mental or emotional state.

[2172] "Activity status" is information that indicates the specific actions and situations that a user is currently engaged in.

[2173] A "generative artificial intelligence model" is an artificial intelligence algorithm that makes predictions and recommendations based on given data.

[2174] "Music provision means" refers to a function that provides music to users using a generative artificial intelligence model.

[2175] "Interaction means" is a function for communicating with users in real time and updating music playlists.

[2176] The "profile collection means" is a means for acquiring the user's current mood, activity status, and emotions.

[2177] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[2178] "Emotion recognition means" refers to a means for monitoring and recognizing a user's emotions and generating emotion data.

[2179] "Data" refers to various information, including information about a user's mood, activities, preferences, emotions, etc.

[2180] "Feedback" refers to opinions and evaluations provided by users to the system.

[2181] "Real-time" refers to actions or processes that respond immediately and without delay.

[2182] The present invention relates to a system for generating music playlists based on a user's preferences and current mood and activity status, which recognizes and reflects the user's current emotions in the music playlist, and provides a personalized music experience through real-time feedback and interaction.

[2183] User profile collection and emotion recognition

[2184] The user begins using the system by launching a music streaming application and logging in. The device asks the user about their current mood and activity (e.g., studying, exercising, or just wanting to relax). The user enters this information into the device, which then collects it. An emotion engine installed on the device recognizes and collects the user's emotions. The emotion engine uses tools such as Affectiva or Microsoft Emotion API. The device retrieves past music playback history from local storage and sends it to the server.

[2185] Data analysis and playlist generation

[2186] The server receives the user's current mood, activity, emotion, and playback history data. The server loads a generative artificial intelligence model and inputs the user's profile data into the model. This model uses GPT-3 or BERT. The server uses this data to generate a music playlist that best suits the user's current situation and emotion. The generated playlist is structured in JSON format and sent to the user's device.

[2187] Real-time playlist updates

[2188] The device displays the generated playlist to the user. As the user plays songs from the playlist, they can enter their feedback about the songs into the device. The emotion engine continues to monitor the user's emotions and collect emotion data. The device sends the user's feedback and emotion data to the server. The server regenerates or updates the playlist based on the feedback and emotion data. The updated playlist is again sent to the user's device in JSON format and displayed.

[2189] User interaction

[2190] Interaction between the user and the system begins when the user makes a request or question to the AI ​​DJ. For example, the user enters a question into the input form, such as "What song do you recommend next?" The device sends this request or question to the server. The server uses the AI ​​DJ model and emotional data to generate an appropriate response and sends it to the user's device. The device then displays or plays the AI ​​DJ's response to the user.

[2191] Specific examples

[2192] The user opens the app and enters their login details to access their account.

[2193] The app prompts you to enter information by popping up questions such as "How are you feeling right now?" and "What are you doing now?"

[2194] The user inputs information such as "I want to relax" or "I'm exercising."

[2195] The emotion engine takes a picture of the user's face with a camera, analyzes their facial expressions, and generates emotion data.

[2196] An instance of the AI ​​model is generated and profile information such as "relaxed," "exercising," and "happy" is provided as input parameters.

[2197] The AI ​​model selects appropriate songs based on the specified profile and emotional data to create playlists such as "Lo-fi Hip Hop" or "House Music."

[2198] The generated playlist is structured in JSON format and sent to the user's device.

[2199] The user types a question into the input form: "What song do you recommend next?"

[2200] The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next song we recommend to help you relax is XX."

[2201] This system allows users to get the best possible music experience based on their preferences and emotional information, and allows for a more personalized music experience through real-time feedback and interaction.

[2202] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2203] Step 1:

[2204] A user launches a music streaming application and logs in.

[2205] Specific operation: The user taps the app to launch it, and the login screen appears. The user enters their username and password and presses the "Login" button.

[2206] Input: Username, Password

[2207] Output: Login session established, user information obtained

[2208] Step 2:

[2209] The device displays a prompt asking the user about their current mood and activity.

[2210] What it does: The app displays a pop-up window, offering the user options such as "How are you feeling right now?" and "What are you currently doing?", as well as a text input field.

[2211] Input: None

[2212] Output: Questions about mood and activity

[2213] Step 3:

[2214] The user inputs their current mood and activity into the device, and the device collects that information.

[2215] Specific operation: The user enters information such as "I want to relax" or "I'm exercising" and presses the "Send" button. The device saves this information in local storage.

[2216] Input: User's mood, activity information

[2217] Output: Mood and activity information stored in local storage

[2218] Step 4:

[2219] The emotion engine installed in the device recognizes and collects the user's emotions.

[2220] How it works: The device's camera captures the user's face, and the facial expression analysis algorithm generates emotional data such as happiness, sadness, and stress, which is then stored in local storage.

[2221] Input: User's face image

[2222] Output: Parsed emotion data

[2223] Step 5:

[2224] The device retrieves past music playback history from local storage and sends it to the server.

[2225] What it does: The device reads past playback history from local storage and sends it to the server as a JSON-formatted data packet along with current mood, activity, and emotion data.

[2226] Input: Playback history, current mood, activity, and emotion data in local storage

[2227] Output: Data packet sent to the server

[2228] Step 6:

[2229] The server receives the user's current mood, activity, emotion, and playback history data.

[2230] Specific operation: The server stores data packets received via the API endpoint in a database.

[2231] Input: User mood, activity, emotion, and playback history data

[2232] Output: User data stored in the database

[2233] Step 7:

[2234] The server loads a generative artificial intelligence model and inputs the user's profile data into the model.

[2235] What it does: The server loads an instance of a generative AI model (e.g., GPT-3) into memory and provides the profile data as input parameters.

[2236] Input: User profile data

[2237] Output: Profile data input to the model

[2238] Step 8:

[2239] The server uses this data to generate a music playlist that best suits the user's current situation and emotions.

[2240] What it does: The model analyzes the input data and lists the best songs. The playlist is structured in JSON format.

[2241] Input: Profile data input to the model

[2242] Output: Structured playlist data

[2243] Step 9:

[2244] The server transmits the generated playlist to the user's terminal.

[2245] Specific operation: Playlist data is sent to the user's device via API, and the device saves the received data.

[2246] Input: Structured playlist data

[2247] Output: Playlist data sent to the user's device

[2248] Step 10:

[2249] The terminal displays the generated playlist to the user.

[2250] Specific operation: Based on the received playlist data, the app displays a song list on the screen.

[2251] Input: Playlist data received from the server

[2252] Output: A list of songs displayed on the screen

[2253] Step 11:

[2254] The user plays a song from the playlist.

[2255] Specific behavior: When the user selects a song and presses the "Play" button, the music will play.

[2256] Input: User song selection

[2257] Output: Played music

[2258] Step 12:

[2259] The user inputs feedback about the song into the terminal, and the emotion engine continuously monitors the user's emotions.

[2260] Specific operation: The user inputs feedback such as "I like this song" or "I want to hear the next song." The device's camera continuously analyzes the user's facial expressions and generates emotion data.

[2261] Input: User feedback, face image

[2262] Output: Feedback data, analyzed emotion data

[2263] Step 13:

[2264] The device transmits the user's feedback and emotional data to the server.

[2265] What it does: Sends feedback and emotion data in real time to a server via an API.

[2266] Input: Feedback data, parsed emotion data

[2267] Output: Feedback and emotion data sent to the server

[2268] Step 14:

[2269] The server regenerates or updates the playlist based on the feedback and emotional data.

[2270] What happens: The server re-runs the AI ​​model based on the new data and generates an updated playlist.

[2271] Input: Feedback and emotion data

[2272] Output: Regenerated or updated playlist data

[2273] Step 15:

[2274] The server sends the updated playlist to the user's terminal.

[2275] Specific behavior: The new playlist data is sent in JSON format to the user's device, and the device displays the playlist again.

[2276] Input: Updated playlist data

[2277] Output: Updated playlist data sent to the user device

[2278] Step 16:

[2279] Users make requests and ask questions to the AI ​​DJ.

[2280] What happens: A user types a question into a text input form, such as "What's the next song you recommend?"

[2281] Input: User request or question

[2282] Output: The input request or question

[2283] Step 17:

[2284] The terminal transmits the user's requests and questions to the server.

[2285] Specific operation: The entered message data is sent to the server via the API.

[2286] Input: Request or question data

[2287] Output: Message data sent to the server

[2288] Step 18:

[2289] The server uses the AI ​​DJ model and emotional data to generate appropriate responses.

[2290] Specific operation: The AI ​​DJ analyzes the request content and emotional information and generates a response message such as, "The next recommended song to help you relax is XX."

[2291] Input: Request or question data, sentiment data

[2292] Output: The generated response message

[2293] Step 19:

[2294] The server generates a response and sends it to the user's terminal.

[2295] Specific operation: A response message is sent to the user's device via the API.

[2296] Input: The generated response message

[2297] Output: Response message sent to the user terminal

[2298] Step 20:

[2299] The terminal displays or plays the AI ​​DJ's responses to the user.

[2300] Specific Behavior: The generated response is displayed to the user in text on their screen or played in audio format.

[2301] Input: Response message sent to the user terminal

[2302] Output: The response message displayed or played on the user's terminal.

[2303] This allows users to get the best possible music experience based on their preferences and emotional information, and allows them to enjoy a more personalized music experience through real-time feedback and interaction.

[2304] (Application example 2)

[2305] 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."

[2306] Conventional music streaming services generate playlists based on users' music preferences, but they do not adequately provide a personalized music experience based on users' real-time emotions and activities. It is also difficult to instantly update playlists to reflect user feedback and emotional data in real time. As a result, it is difficult to provide users with the optimal music experience.

[2307] 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 a music providing means having a generative artificial intelligence model that generates a music playlist based on the user's preferences, an interaction means that updates the music playlist in real time through interaction with the user, a profile collecting means that acquires the user's current mood and activity status, an input means that inputs the data collected by the profile collecting means into the generative artificial intelligence model, an emotion recognition means that captures the user's facial expressions and analyzes their emotions, and an emotion input means that inputs the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model. This enables a music experience based on the user's real-time emotions and activities.

[2308] A "music providing means" is a means that includes a generative artificial intelligence model that generates a music playlist based on user preferences.

[2309] The "interaction means" is a means for updating the music playlist in real time by interacting with the user.

[2310] The "profile collection means" is a means for acquiring the user's current mood and activity status.

[2311] The "input means" is a means for inputting the data collected by the profile collection means into the generative artificial intelligence model.

[2312] An "emotion recognition means" is a means for capturing a user's facial expressions and analyzing their emotions.

[2313] The "emotion input means" is a means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[2314] A "generative AI model" is an AI model equipped with an algorithm that generates playlists based on user preferences and updates them based on real-time data.

[2315] A "music playlist" is a list of music generated based on a user's preferences and emotions.

[2316] This invention provides a system that generates a music playlist based on a user's preferences and updates the playlist in real time. The system uses emotional data obtained from the user's current mood, activity status, and facial expressions to provide appropriate music content in a personalized manner. A specific embodiment of this invention will be described below.

[2317] Hardware and software used

[2318] Hardware:

[2319] Webcam: To capture the user's facial expressions

[2320] Smartphones and tablets: To interact with the user and play music

[2321] software:

[2322] OpenCV: A library for capturing and analyzing facial expressions

[2323] TensorFlow: A library for running emotion recognition models

[2324] requests: An HTTP request library for sending data to servers and retrieving music playlists.

[2325] System configuration

[2326] 1. Generative AI model that generates music playlists based on user preferences: Equipped with an AI model that generates optimal playlists based on the music a user has played in the past and their profile information.

[2327] 2. Interaction: A means to interact with the user in real time and update the music playlist. The user makes requests via text or voice input.

[2328] 3. Profile collection means: A means of obtaining a user's current mood and activity status, including information the user enters into the device and daily activity information.

[2329] 4. Input means: A means for inputting data collected by the profile collection means into the generative artificial intelligence model.

[2330] 5. Emotion Recognition: A means of capturing the user's facial expressions and analyzing their emotions. Specifically, facial images taken with a webcam are preprocessed using OpenCV and then analyzed using TensorFlow's emotion recognition model.

[2331] 6. Emotion input means: A means for inputting the emotion data analyzed by the emotion recognition means into the generative artificial intelligence model.

[2332] Process Overview

[2333] The server first generates a music playlist based on the user's preferences. This process is based on past music playback history, current mood, activity status, and emotional data. The webcam captures the user's facial expressions and analyzes their emotions. This analysis, along with their profile information, is then fed into a generative artificial intelligence model to provide the optimal music playlist.

[2334] Examples of concrete examples and prompts

[2335] Examples:

[2336] The scenario imagines a user wearing smart glasses and a camera capturing their facial expressions.

[2337] The emotion recognition model detects when the user is "happy."

[2338] The server then sends the user a playlist of "relaxing music that matches a happy mood" along with the emotional data.

[2339] Example prompt sentence:

[2340] User ID: user12345, current emotion: happy

[2341] Recommend the best music content for this user.

[2342] This system enables a music experience based on the user's real-time emotions and activities, providing a deeper, more personalized entertainment experience.

[2343] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2344] Step 1:

[2345] A user launches a music streaming application and logs in. The user opens the app, enters their login information, and accesses their account. During this process, a set of initial data is collected, allowing the system to obtain the user's identity and profile information.

[2346] Step 2:

[2347] The device asks the user about their current mood and activity. Specifically, the app displays a pop-up asking questions such as "How are you feeling right now?" and "What are you currently doing?" to prompt the user for input. In response, the user provides information such as "I want to relax" or "I'm exercising." This information is collected by the profile collection means.

[2348] Step 3:

[2349] The device uses a webcam to capture the user's facial expressions. The emotion recognition unit preprocesses this facial image and converts it to grayscale. It then uses a face detection model to identify the face's location. The identified facial features are input into TensorFlow's emotion recognition model to analyze the user's emotions. The analysis results in emotional data such as "happiness," "anger," or "surprise."

[2350] Step 4:

[2351] The device sends the collected mood and activity data, as well as the analyzed emotional data, to the server, which then structures the data along with the profile information and sends it to the server, which receives it and prepares the data for real-time analysis.

[2352] Step 5:

[2353] The server inputs the received data into a generative AI model. The model is fed with the user's current mood, activity, emotions, and past music playback history. The generative AI model generates an optimal music playlist based on this data. The model analyzes the data and selects music to generate an optimal song list.

[2354] Step 6:

[2355] The server sends the generated music playlist to the device. The playlist is structured in JSON format and sent to the user's device. The device receives the playlist and displays it to the user.

[2356] Step 7:

[2357] The user plays songs from the generated playlist, and the device plays the selected songs and collects user feedback during the process. Feedback information is collected through simple form-filling and sentiment analysis.

[2358] Step 8:

[2359] The device sends the collected feedback and real-time emotional data to the server, which then inputs it back into the generative AI model to regenerate or update the playlist. The updated playlist is then sent back to the device and displayed to the user. This cycle creates a dynamic music experience for the user.

[2360] 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.

[2361] 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.

[2362] 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.

[2363] 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.

[2364] 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.

[2365] 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.

[2366] 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).

[2367] 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.

[2368] 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."

[2369] 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.

[2370] 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).

[2371] 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.

[2372] 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.

[2373] 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.

[2374] 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.

[2375] 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.

[2376] 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.

[2377] 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.

[2378] 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.

[2379] 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.

[2380] 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.

[2381] The following is further disclosed regarding the above embodiment.

[2382] (Claim 1)

[2383] a music providing means having a generative artificial intelligence model that generates a music playlist based on a user's preferences;

[2384] an interaction means for updating the music playlist in real time by interacting with the user;

[2385] a profile collection means for acquiring a user's current mood and activity status;

[2386] an input means for inputting the data collected by the profile collection means into the generative artificial intelligence model;

[2387] A system including:

[2388] (Claim 2)

[2389] 10. The system of claim 1, wherein the generative artificial intelligence model comprises an algorithm that analyzes a user's past music playback history and recommends new music based on the user's preferences.

[2390] (Claim 3)

[2391] 10. The system of claim 1, wherein the means for interacting with the generative artificial intelligence model accepts requests from the user via text and voice input.

[2392] "Example 1"

[2393] (Claim 1)

[2394] a music providing means having a generative artificial intelligence model that generates a music playlist based on a user's preferences and current mood and activity;

[2395] an interaction means for updating the music playlist in real time by interacting with the user;

[2396] a profile collection means for acquiring a user's current mood and activity status;

[2397] an input means for inputting the data collected by the profile collection means into the generative artificial intelligence model;

[2398] A system including:

[2399] (Claim 2)

[2400] 10. The system of claim 1, wherein the generative artificial intelligence model comprises an algorithm that analyzes a user's past music playback history and recommends new music based on the user's preferences.

[2401] (Claim 3)

[2402] 10. The system of claim 1, wherein the means for interacting with the generative artificial intelligence model accepts requests from the user via text and voice input.

[2403] "Application Example 1"

[2404] (Claim 1)

[2405] a recommendation means having a generative artificial intelligence model that generates optimal recommendations based on user preferences;

[2406] An interaction means for updating recommendations in real time by interacting with the user;

[2407] a profile collection means for acquiring a user's current mood and activity status;

[2408] an input means for inputting the data collected by the profile collection means into the generative artificial intelligence model;

[2409] A system including:

[2410] (Claim 2)

[2411] The system of claim 1, wherein the generative artificial intelligence model comprises an algorithm that analyzes a user's past ordering history and recommends new dishes and restaurants based on the user's preferences.

[2412] (Claim 3)

[2413] 10. The system of claim 1, wherein the means for interacting with the generative artificial intelligence model accepts requests from the user via text and voice input.

[2414] "Example 2: Combining E...

Claims

1. a music providing means having a generative artificial intelligence model that generates a music playlist based on a user's preferences; an interaction means for updating the music playlist in real time by interacting with the user; a profile collection means for acquiring a user's current mood and activity status; an input means for inputting the data collected by the profile collection means into the generative artificial intelligence model; A system including:

2. 10. The system of claim 1, wherein the generative artificial intelligence model comprises an algorithm that analyzes a user's past music playing history and recommends new music based on the user's preferences.

3. 10. The system of claim 1, wherein the means for interacting with the generative artificial intelligence model accepts requests from the user via text and voice input.

Images