System

Abstract

A system is provided.SOLUTION: A system includes a means for receiving and analyzing a game video in real time, a means for extracting information on a specific target player on the basis of an analysis result, a means for generating a live commentary on the specific target player on the basis of the extracted information, and a means for transmitting the generated live commentary to a user terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional live sports broadcasts often contain a lot of information other than the viewer's target athlete (hereinafter referred to as the "target athlete"), making it difficult for viewers to understand the performance status of their favorite athlete in real time. Furthermore, there is a lack of information available that focuses on the target athlete, which hinders viewers' satisfaction with their favorite athlete's activities. The present invention aims to solve these problems. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for receiving and analyzing game footage in real time, a means for extracting information about a specific target player based on the analysis results, a means for generating a commentary about the specific target player based on the extracted information, and a means for transmitting the generated commentary to a user terminal.The system also includes a means for a user to select a specific target player, a means for transmitting the selected information to a server, a means for receiving and displaying the commentary from the server, a means for generating a natural language commentary based on the extracted information, and a means for converting the generated natural language commentary into audio data, allowing users to receive real-time commentary tailored to their favorite player, thereby improving their satisfaction with their favorite player activities.

[0006] "Game footage" is video data showing a competitive event such as a sport.

[0007] "Real-time" refers to the immediate processing and display of ongoing events.

[0008] "Means of analysis" refers to the functions and methods for analyzing input data and extracting meaningful information from it.

[0009] A "specific target athlete" is a specific athlete that a user chooses to support.

[0010] "Means of extracting information" refers to the processes and functions for extracting necessary information from vast amounts of data.

[0011] The "means for generating commentary" refers to the process or function for verbally explaining the sports situation based on the extracted information.

[0012] A "user terminal" is an electronic device used by a user, such as a computer, smartphone, or tablet.

[0013] A "transmission medium" is a process or technique for transferring data from one point to another.

[0014] A "means of choice" is a process or function for selecting a specific one from multiple options.

[0015] A "server" is a large computer system that provides services to other computers over a network.

[0016] "Means for receiving" refers to the processes and functions for receiving data sent from outside.

[0017] "Natural language commentary" refers to explaining the progress of a game using language that humans use on a daily basis.

[0018] "Audio data" means data for recording, storing, and transmitting audio information in digital form. [Brief explanation of the drawings]

[0019] [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

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete). Below, we will explain how the program of this system works, using concrete examples.

[0041] Setting favorite players on user devices

[0042] To use the system, a user first selects a specific target player to support. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the application on their smartphone and selects "Player A" from the player list.

[0043] The user terminal acquires this selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0044] Real-time analysis of game footage by server

[0045] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, it detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0046] Identifying information about a specific target player by the server

[0047] The server extracts information related to the specific target player from the analysis results. In this process, the server precisely records the specific target player's important actions (e.g., dribbling, shooting, passing), their position information, and the surrounding situation. This allows users to obtain the information they want in real time.

[0048] Server-based commentary generation

[0049] Based on the extracted information, the server generates a commentary in natural language. The generated commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0050] Transmission of commentary generated by the server and display on the user's device

[0051] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to check the latest status of their favorite player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0052] This system allows users to obtain timely information specific to their favorite athletes, eliminating the frustrations of conventional live sports broadcasts. In this way, the present invention significantly improves the user's sports viewing experience.

[0053] The processing flow will be explained below.

[0054] Step 1:

[0055] The user launches the application and selects a specific target player. The user selects their favorite target player from the player list in the application and enters the selection information into the device.

[0056] Step 2:

[0057] The terminal sends the selection information to the server. The terminal generates a request including the identification information (e.g., player ID, name) of the specific target player selected by the user and sends it to the server.

[0058] Step 3:

[0059] The server receives live-streamed game footage in real time. The server continuously receives the game footage data stream and stores it in a buffer.

[0060] Step 4:

[0061] The server analyzes the game footage frame by frame, and the server's generative AI model processes the received footage frame by frame to identify the positions and movements of players in each frame.

[0062] Step 5:

[0063] The server extracts information about the specific target player. The server extracts information related to the specific target player selected by the user (e.g., position data, ball possession status, actions) from the analyzed video data.

[0064] Step 6:

[0065] The server generates a commentary based on the extracted information. The server generates a commentary in natural language using the information on the specific target player, and then converts the commentary into audio data using speech synthesis technology.

[0066] Step 7:

[0067] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0068] Step 8:

[0069] The terminal displays and plays the commentary received. The terminal displays the text commentary received from the server on the screen and plays the audio commentary, allowing the user to understand the real-time situation of a specific target player.

[0070] Example 1

[0071] 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."

[0072] Conventional sports viewing systems have difficulty obtaining real-time information about specific athletes that users are interested in. Furthermore, conventional systems lack sufficient video analysis and commentary generation for specific athletes selected by users, resulting in an unsatisfactory sports viewing experience for users. Furthermore, the mechanisms for providing generated information to user devices in real time are also inadequate.

[0073] 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.

[0074] In this invention, the server includes means for receiving and analyzing game footage in real time, means for identifying specific target players based on the analysis results, means for extracting information about the identified specific target players, means for generating a commentary in natural language based on the extracted information, means for converting the generated natural language commentary into audio data, and means for transmitting the generated commentary to a user terminal, thereby enabling users to obtain information about specific players in real time and significantly improving their viewing experience.

[0075] "Game video" refers to video data that records the state of a sports game.

[0076] "Means for receiving and analyzing in real time" refers to a combination of hardware and software for receiving game footage in real time via a network and sequentially analyzing the footage.

[0077] The "means for identifying a specific target player based on the analysis results" refers to an algorithm and program for identifying a specific player from the analyzed game video data.

[0078] The "means for extracting information about the identified specific target player" is a program for extracting related information such as the position information and movements of the specific player, and their positional relationship with other players, from the analysis results.

[0079] The "means for generating commentary in natural language" refers to an algorithm and program for generating commentary in text format based on the extracted information.

[0080] The "means for converting the commentary in natural language into voice data" is a voice synthesis technology for converting the generated commentary in text format into voice data.

[0081] "Means for transmitting to user terminal" refers to network technologies and protocols for transmitting the generated commentary to the user's device in real time.

[0082] The "means for the user to select a specific target player" refers to an interface for the user to select a favorite player and a program for confirming the selection.

[0083] The "means for transmitting selected information to the server" refers to a network technology for transmitting information about the specific target player selected by the user to the server.

[0084] The "means for generating a request including identification information of a selected specific target player" is a program for generating a data request including identification information about the selected player.

[0085] The "means for transmitting the generated request to the server" refers to a network technology for transmitting the generated request to the server.

[0086] The "means for receiving and displaying the live commentary from the server" refers to a program and interface for receiving the live commentary sent from the server at the user terminal and displaying it on the screen.

[0087] "Means for generating information in natural language using a generative AI model" means a program that uses a generative AI model (e.g., GPT, BERT) to generate commentary in natural language based on extracted information.

[0088] The present invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system combines a user terminal, a server, and a generative AI model to analyze game footage in real time and provide information.

[0089] Users access the system through a smartphone application. When the application is launched, the user can access an interface for selecting a favorite player. When the user selects a specific player (e.g., "Player A") from the player list and confirms the selection, the user's device acquires this information. The device generates a request including the identification information (player ID) of the selected specific target player and sends it to the server. This request is sent using the HTTP POST method.

[0090] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API) and uses a GPU to perform high-speed image analysis. The software used includes generative AI models (e.g., YOLO and OpenPose). Using these models, the server analyzes the game footage frame by frame to detect the positions and movements of specific target players. For example, in a soccer game, the analysis identifies the position and actions (e.g., dribbling, shooting, passing) of "Player A."

[0091] The server extracts information related to "Player A" from the analysis results. During this extraction process, "Player A's" location information, actions, and surrounding circumstances are recorded in detail. This information is used to grasp the data the user wants to know in real time.

[0092] The server generates a commentary in natural language based on the extracted information. A generative AI model (e.g., GPT-3 or BERT) is used for natural language generation. By inputting a prompt sentence such as "Player A is dribbling," the generative AI model generates specific commentary text such as "Player A is dribbling forward, dodging the defender!" This text commentary is then converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly).

[0093] The generated commentary (text and audio data) is sent from the server to the user's device. WebSocket and HTTP / 2 protocols are used to transmit data in real time. The user's device immediately processes the received commentary data. The received commentary text is displayed on the display, and the audio commentary is played through the speaker.

[0094] As a concrete example, consider the case where a user selects a soccer player, "Player A," as their favorite player. The user selects "Player A" using a smartphone app, and that information is sent to the server. The server receives the game footage, analyzes it using a generative AI model, and identifies the scene where "Player A" is dribbling in front of the goal. As a result, a commentary such as "Player A is approaching the goal!" is generated and immediately sent to the user's smartphone. The user's smartphone displays this commentary and simultaneously plays it back as audio.

[0095] Examples of prompts:

[0096] "Player A starts dribbling at 15:42"

[0097] "Player A is dribbling forward past the defender!"

[0098] As described above, the system of the present invention allows users to obtain real-time information specific to a particular athlete, greatly enhancing the sports viewing experience.

[0099] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0100] Step 1:

[0101] User-defined favorite players

[0102] The user launches the smartphone application and sets their favorite player. Specifically, they select a specific target player (for example, "Player A") from the player list in the application and confirm the selection. The input data is the player list and player ID, and the output data is the identification information (player ID) of the selected player. The user selects "Player A," and that ID is stored within the application.

[0103] Step 2:

[0104] Sending a request from the device to the server

[0105] The terminal generates a request including the identification information (player ID) of the specific target player selected by the user. This request is sent to the server using the HTTP POST method. The input data is the player ID, and the output data is the request sent to the server. The terminal converts the player ID into a JSON object and sends it to the URL https: / / api.example.com / select-player.

[0106] Step 3:

[0107] Real-time analysis of game footage by server

[0108] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API). It uses a GPU to process the video data frame by frame and analyzes player movements using a generative AI model (e.g., YOLO or OpenPose). The input data is the live stream video, and the output data is player position and movement information. The server analyzes the game footage frame by frame to identify the position and movement of "Player A."

[0109] Step 4:

[0110] Server extracts information about specific target players

[0111] The server extracts information related to a specific target player ("Player A") from the analysis results. The input data is the frame analysis results, and the output data is "Player A's" position information, actions, and surrounding circumstances. The server records this information in a database, allowing users to grasp the information they want to know in real time.

[0112] Step 5:

[0113] Server-based commentary generation

[0114] The server generates commentary in natural language based on the extracted information. Using a generative AI model (e.g., GPT-3 or BERT), a prompt sentence is input and specific commentary text is generated. The input data is the extracted player information, and the output data is commentary text in natural language. For example, the server generates commentary text such as, "Player A is dribbling forward, dodging the defender!"

[0115] Step 6:

[0116] Live audio conversion by the server

[0117] The generated commentary text is converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The input data is the commentary text, and the output data is audio data. The server converts the generated commentary text into an audio file.

[0118] Step 7:

[0119] Live data transmission from the server to the device

[0120] The server transmits the generated commentary text and audio data to the user terminal in real time. The input data is the commentary text and audio data, and the output data is the transmitted commentary data. This transmission is performed using the WebSocket or HTTP / 2 protocol.

[0121] Step 8:

[0122] Live display and audio playback on the device

[0123] The device processes the commentary data received from the server, displays the text on the display, and plays the audio from the speaker. The input data is the commentary text and audio data, and the output data is the displayed text and the played audio. For example, the commentary text "Player A is closing in on the goal!" is displayed on the screen, and the content is played back as audio.

[0124] (Application example 1)

[0125] 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."

[0126] In conventional sports commentary, it was difficult to obtain detailed information about specific players in real time. In addition, users were unable to receive timely information specific to their favorite players, and were only provided with general information that was in line with the progress of the entire game. This limited users' sports viewing experience, preventing them from enjoying the excitement of focusing on the play of their favorite players in real time.

[0127] 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.

[0128] In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about specific target players based on the analysis results, means for generating commentary about the specific target players based on the extracted information, means for transmitting the generated commentary to a user terminal, and means for displaying and playing back the received commentary on the user terminal as audio. This allows users to receive detailed information about the play of their favorite players in real time, providing a more exciting sports viewing experience through both their eyes and their ears.

[0129] "Game footage" is video data captured during a sporting event.

[0130] "Receiving in real time" means obtaining data without delay while the match is in progress.

[0131] "Means for analysis" refers to a device or program that can extract the movements and positional information of specific players from the received game footage.

[0132] A "specific target athlete" is a specific athlete designated by the user.

[0133] "Means for extracting information" refers to devices or programs that extract data related to specific target players from the analysis results of game footage.

[0134] "Means for generating commentary" refers to a device or program that creates commentary in natural language from the extracted player information.

[0135] A "user terminal" is an electronic device used by a user, and examples include smartphones and tablets.

[0136] The "transmitting means" refers to a communication device or program for delivering the generated commentary to the user terminal.

[0137] "Means for displaying and playing audio" refers to a device or program that displays the received commentary on a screen and plays the audio through speakers or headphones.

[0138] "User-selected means" refers to the interface or operation method that allows the user to specify their favorite player.

[0139] "Means for sending to server" refers to a communication device or program for sending the information selected by the user to the central server.

[0140] "Natural language commentary" is commentary that expresses information about a specific player in natural language that is easy for people to understand.

[0141] "Means for converting into audio data" refers to a device or program that converts the generated natural language text into an audio file.

[0142] This invention is a system that provides real-time commentary specific to a specific athlete. This system provides users with detailed information such as the movements and positions of the specific athlete in real time, thereby improving the sports viewing experience. Specific embodiments of this system are described below.

[0143] System Configuration

[0144] The server includes the following means:

[0145] 1. Means of receiving and analyzing game footage in real time:

[0146] The server receives live-streamed game footage in real time and uses deep learning models such as TensorFlow for video analysis.

[0147] 2. Methods for extracting information on specific target players based on the analysis results:

[0148] The received video is analyzed frame by frame to extract the movement and position information of specific target players, which includes player identification and action recognition.

[0149] 3. A means for generating commentary about a specific target player based on the extracted information:

[0150] Based on the extracted information, a live commentary text is generated using natural language processing technology, using generative AI models such as Transformers for natural language generation.

[0151] 4. Means for transmitting the generated commentary to the user terminal:

[0152] The generated text and voice data are sent to the user's device using real-time data transfer technology (e.g., WebSocket) over the Internet.

[0153] The user terminal includes the following means:

[0154] 1. Means for displaying and playing back received commentary:

[0155] The user device (e.g., a smartphone) displays the received commentary text on its screen and plays the audio data through speakers or headphones using a voice playback and speech synthesis library such as pygame or gTTS.

[0156] 2. How users select specific target players:

[0157] The user selects their favorite player through the application, and the selection information is sent to the server, where the player is set as a specific target player.

[0158] Specific examples

[0159] Specific examples are given below.

[0160] Let's take the example of a scenario where a user is using "Favorite Player Live Assist" on their smartphone to watch a soccer match. The user selects Player A from the application's list and sets him as the specific target player. Meanwhile, the server analyzes the game footage in real time and identifies Player A's movements. For example, when it captures a scene where Player A has the ball and begins to dribble, it generates a commentary such as, "Player A is dribbling forward, dodging the defender." This commentary is immediately sent to the smartphone and displayed on the screen, while also being played back as audio using speech synthesis technology.

[0161] Example prompts for generative AI models

[0162] "Player A is dribbling forward past the defender. Please provide a commentary of the game situation."

[0163] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0164] Step 1:

[0165] The user selects a specific target player

[0166] The user launches the application and selects a specific target player from a list. The user selects their favorite player from the player list (loaded from the database) by tapping on it. As input, the player's identification information is generated and sent to the server. As output, the player ID is selected and sent to the server.

[0167] Step 2:

[0168] Send the selected information to the server

[0169] The terminal sends the identification information of the specific target player selected by the user to the server. The server receives the information and sets the specified player as the specific target player. The player ID is sent from the terminal as input, and the player information is saved in the database in the server as output.

[0170] Step 3:

[0171] The server receives and analyzes the match footage in real time.

[0172] The server receives live-streamed game footage in real time and analyzes it using a deep learning model (such as TensorFlow). Live video data is given as input, and frame-by-frame analysis results are output. Specifically, player movements and position information are identified for each frame.

[0173] Step 4:

[0174] Extract information on specific target players based on the analysis results

[0175] The server extracts information related to a specific target player from the analysis results. The analyzed data of all players is given as input, and data showing the actions and positions of the specific target player is obtained as output. Specifically, the player's actions (e.g., dribbling, passing, shooting) and current position information are extracted.

[0176] Step 5:

[0177] Generate commentary based on the extracted information

[0178] Based on the extracted information, the server uses a generative AI model (e.g., GPT-3) to generate commentary text in natural language. The input is given as movement data of a specific target player, and the output is generated commentary text. Specifically, the generative AI creates text using the prompt sentence, "Player A is dribbling forward, dodging the defender. Please provide commentary on the situation in the game."

[0179] Step 6:

[0180] Send the generated commentary to the user's device

[0181] The server transmits the generated commentary text and audio data to the user terminal. The commentary text and audio files are stored in the server as input, and are transmitted to the user terminal as output. Specifically, the data is transmitted in real time via the Internet.

[0182] Step 7:

[0183] The user device displays the commentary and plays it back with audio.

[0184] The user device displays the received commentary text on the screen and plays the audio data through speakers or headphones. The input is the received text and audio file, and the output is the display of text on the screen and audio playback. The user can follow the movements of a specific player in real time through both sight and sound.

[0185] 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.

[0186] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete), and further combines it with an emotion engine that recognizes the user's emotions. How the program of this system works will be explained using concrete examples.

[0187] Setting favorite players on user devices

[0188] To use the system, a user must first select a specific target player. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the smartphone app and selects "Player A" from the player list.

[0189] The user terminal acquires the selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0190] Real-time analysis of game footage by server

[0191] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0192] Identifying information about a specific target player by the server

[0193] The server extracts information related to the specific target player from the analysis results. During this process, the key actions of the specific target player (e.g., dribbling, shooting, passing), their position, and the surrounding situation are precisely recorded. This allows users to obtain the information they want in real time.

[0194] Server-based commentary generation

[0195] Based on the extracted information, the server generates commentary in natural language. This commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0196] Transmission of commentary generated by the server and display on the user's device

[0197] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to keep track of the latest status of a specific target player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the same commentary in audio.

[0198] Recognizing user emotions with an emotion engine

[0199] The user device is equipped with an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the user's facial expressions and voice tone in real time to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[0200] Adjusting commentary based on emotion information

[0201] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to make it even more exciting. A rich commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!"

[0202] This system allows users to understand the status of specific athletes in real time and receive commentary that is customized to their individual emotions, eliminating frustrations with traditional sports commentary and significantly improving the user's sports viewing experience.

[0203] The processing flow will be explained below.

[0204] Step 1:

[0205] The user launches the application and selects a specific player. The user selects "Player A" from the player list in the app.

[0206] Step 2:

[0207] The terminal transmits the selection information to the server, and the terminal generates a request including the identification information of the specific target player and transmits it to the server.

[0208] Step 3:

[0209] The server receives live-streamed game footage in real time and stores the game footage data stream in a buffer.

[0210] Step 4:

[0211] The server analyzes the game footage frame by frame, and the server's generative AI model processes each frame to identify the players' positions and movements.

[0212] Step 5:

[0213] The server extracts information about the specific target player, and extracts information related to the specific target player from the analysis results (e.g., position data, ball possession status, actions).

[0214] Step 6:

[0215] The server generates commentary based on the extracted information, using natural language generation technology to generate commentary such as, "Player A is dribbling forward, dodging the defender!"

[0216] Step 7:

[0217] The server converts the generated commentary into audio data, and the generated text commentary is converted into audio data using voice synthesis technology.

[0218] Step 8:

[0219] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0220] Step 9:

[0221] The device displays and plays the commentary it receives. The device displays the received text commentary on the screen and plays the received audio commentary.

[0222] Step 10:

[0223] The device's emotion engine recognizes the user's emotions by analyzing the user's facial expressions and tone of voice to identify the user's emotional state.

[0224] Step 11:

[0225] The device transmits emotion information to the server. The emotion engine transmits the user's emotion information identified by the device to the server.

[0226] Step 12:

[0227] The server adjusts the commentary based on the emotional information. The server receives the user's emotional state and dynamically adjusts the commentary content based on the information.

[0228] Step 13:

[0229] The server transmits the adjusted commentary to the user terminal, and the server transmits the adjusted commentary data to the user terminal again.

[0230] Step 14:

[0231] The terminal displays and plays the adjusted commentary. The terminal displays the received adjusted commentary on the screen and plays it back as audio.

[0232] Example 2

[0233] 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."

[0234] Conventional sports commentary systems make it difficult for users to receive real-time information about specific athletes, and they are unable to provide customized commentary that takes into account the user's emotional state. This results in a standardized viewing experience for users, and makes it difficult to meet individual needs.

[0235] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for recognizing the user's emotional state, means for dynamically adjusting the commentary content based on the recognized emotional information, and means for transmitting the generated commentary to the user terminal. This not only enables the user to receive detailed real-time information about a specific player, but also makes it possible to provide a commentary customized according to the user's emotions.

[0236] "Game footage" refers to video data that records the progress of a sports game in real time.

[0237] The "analysis means" refers to a technology or device that analyzes the input game video and extracts important information from each frame.

[0238] "Specific target athlete" refers to a specific athlete selected by the user.

[0239] "Means for extracting information" refers to a technique or method for extracting data (position information, movements, etc.) related to a specific target player from the analysis results.

[0240] The "means for generating commentary" is a technique or method for describing the actions of a specific target player and the situation of the game in natural language based on the extracted data.

[0241] "User's emotional state" refers to the emotional state obtained by analyzing the user's facial expression, voice tone, and other biometric information.

[0242] The "means for recognizing emotional states" refers to a technology or device that analyzes the user's facial expressions and tone of voice to identify emotions.

[0243] A "means for dynamically adjusting commentary content" is a technique or method for changing commentary content in real time based on recognized emotion information.

[0244] "User terminal" refers to a computer, smartphone, tablet, or other mobile device used by a User.

[0245] The "generated commentary" is text and audio data of a narration generated based on the progress of the match.

[0246] The "transmitting means" refers to a technique or method for transmitting the generated commentary to a user terminal.

[0247] This invention is a system that allows users to receive real-time commentary on specific athletes (hereinafter referred to as "specific target athletes") and further customizes the commentary content by recognizing the user's emotions. The program processing of this system is explained in detail below.

[0248] To use the system, a user must first select a specific target player. The user launches the smartphone app and selects a specific target player from the player list. For example, the user selects "Player A." At this time, the smartphone generates a request including the player's identification information and sends it to the server.

[0249] The server receives this request and uses a generative AI model such as YOLOv5 to analyze each frame of the live-streamed game footage in real time, identifying the position and movement of specific target players and extracting important actions (e.g., dribbling, shooting, passing).

[0250] Based on the extracted information, the server generates commentary for the specific target player using a natural language generation model such as GPT-3. For example, a commentary such as "Player A is dribbling forward, dodging the defender!" is generated. This commentary is then converted into audio data using voice synthesis technology such as Google Cloud Text-to-Speech.

[0251] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays this commentary in real time and plays the audio. For example, the user's smartphone might display the text "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0252] Furthermore, the user device is equipped with an emotion engine that analyzes the user's facial expressions and voice tone in real time to identify their emotional state. The emotion engine uses TensorFlow and other tools to recognize the user's surprise, excitement, joy, etc., and sends this information to the server. The server receives the user's emotional information and, for example, if the user is very excited, adjusts the commentary to make it more exciting. Commentary that corresponds to the user's emotional state is generated, such as "Player A is showing off some amazing dribbling!"

[0253] This system not only allows users to keep up with the latest game progress of specific players in real time, but also allows them to receive commentary that is customized to their emotions at the time. This technology meets individual needs that traditional sports commentary could not meet, significantly improving the user's viewing experience.

[0254] Examples of specific examples and prompts

[0255] For example, if a user "Tanaka" has set a specific target player "Player A" on his smartphone and is watching a soccer game, and Tanaka looks surprised when "Player A" shoots the ball:

[0256] 1. The user device detects Tanaka's surprised expression, and the emotion engine analyzes it.

[0257] 2. The emotion engine sends the recognized emotion information to the server.

[0258] 3. The server adjusts the commentary based on this information, generating a commentary such as "Player A took a great shot!"

[0259] 4. This live commentary will be displayed in real time on Tanaka's smartphone and played back as audio.

[0260] Example prompt sentence:

[0261] "Generate a commentary of a game played by Player A. If the user seems excited, make the commentary even more exciting."

[0262] The above is a specific embodiment of the present invention, which allows users to enjoy detailed game situations of specific target players and commentary customized to their own emotions in real time.

[0263] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0264] Program processing flow

[0265] Step 1: User's Favourite Player

[0266] 1. The user launches the smartphone app and selects a specific target player from the player list.

[0267] Input: User selects a player within the app

[0268] Output: Identification information of the specific target player is generated.

[0269] Specific operation: When the user selects "Player A" on the app screen, the device generates the identification information of the specific target player (e.g., { "player_id": "A123"}).

[0270] 2. The terminal generates a request including the selection information and sends it to the server.

[0271] Input: Identification information of specific target player

[0272] Output: The request sent to the server

[0273] Specific operation: The device sends a request containing the device-generated identification information to the server in the following format: { "action": "select_player", "player_id": "A123"}

[0274] Step 2: Real-time analysis of game footage by the server

[0275] 1. The server receives live footage of the match.

[0276] Input: Live video stream

[0277] Output: Image data for each frame

[0278] Specific operation: The server uses tools such as FFmpeg to divide the live video into frames and generate image data for each frame.

[0279] 2. The server analyzes each frame using a generative AI model to detect the position and movement of specific target players.

[0280] Input: Frame-by-frame image data

[0281] Output: Position and action information of specific target players

[0282] Specific behavior: Using a generative AI model such as YOLOv5, each frame is analyzed to detect the position (e.g., coordinate information) and actions (e.g., dribbling, shooting) of specific target players.

[0283] Step 3: The server identifies information about the specific target player.

[0284] 1. The server extracts important actions of specific target players from the analysis results.

[0285] Input: Detected location and action information

[0286] Output: List of important actions of a specific target player

[0287] Specific actions: Extract the main actions of a specific target player from the analysis results (e.g., { "time": "10:12", "action": "dribble"}) and save them in list format.

[0288] Step 4: Server-generated commentary

[0289] 1. The server generates commentary text based on the extracted information.

[0290] Input: List of important actions for a specific target player

[0291] Output: Commentary text

[0292] Specific Actions: Use generative AI models such as GPT-3 to generate commentary text based on specific player actions and game situations (e.g., "Player A is dribbling forward, dodging the defender!").

[0293] 2. The server converts the generated text into audio data.

[0294] Input: Play-by-play text

[0295] Output: Audio data

[0296] Specific operation: The generated commentary text is converted into audio data using Google Cloud Text-to-Speech or similar.

[0297] Step 5: Transmission of the generated commentary by the server and display on the user's device

[0298] 1. The server sends the generated commentary (text and audio data) to the user terminal.

[0299] Input: Live text and audio data

[0300] Output: Live data sent to the user's device

[0301] Specific operation: A network communication protocol (e.g., HTTP, WebSocket) is used to send live data to the user terminal.

[0302] 2. The device displays the received commentary in real time and plays it back as audio.

[0303] Input: Live data from the server

[0304] Output: Running text displayed on the screen and audio played back

[0305] Specific operation: The received commentary text is displayed on the device screen and the audio data is played back.

[0306] Step 6: Recognizing user emotions with the emotion engine

[0307] 1. The device collects the user's facial expressions and voice tone in real time.

[0308] Input: User's facial and voice data

[0309] Output: Collected real-time emotion data

[0310] Specific operation: The device's camera and microphone are used to collect the user's facial expressions and voice tone in real time.

[0311] 2. The device analyzes the collected emotional data and identifies the user's emotional state.

[0312] Input: Collected emotion data

[0313] Output: Parsed emotional state information

[0314] Specific operation: Analyze the user's emotional state (e.g., surprise, excitement) using an emotion recognition engine such as TensorFlow.

[0315] 3. The device sends the analyzed emotion information to the server.

[0316] Input: Parsed emotional state information

[0317] Output: Emotion information sent to the server

[0318] Specific operation: A network communication protocol is used to transmit emotional state information to the server.

[0319] Step 7: Adjusting commentary based on emotional information

[0320] 1. The server receives the user's emotional information and adjusts the commentary content based on that emotional information.

[0321] Input: User's emotional information

[0322] Output: Adjusted commentary text

[0323] Specific operation: Based on the user's emotional information, the commentary text is regenerated using a generative AI model such as GPT-3 and the content is adjusted.

[0324] 2. The server sends the adjusted commentary back to the user's device.

[0325] Input: Adjusted play-by-play text

[0326] Output: Adjusted live data sent to user device

[0327] Specific operation: The adjusted live data is sent back to the user's terminal, and the terminal displays and plays it.

[0328] The above are the specific processing steps of the system program, and this series of processes enables users to receive real-time information on specific players and customized commentary according to their emotions.

[0329] (Application example 2)

[0330] 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."

[0331] Conventional sports commentary systems lacked commentary tailored to specific athletes, making it difficult to provide information tailored to users' interests. Furthermore, they lacked commentary tailored to the user's emotions, resulting in a limited viewing experience and low satisfaction. Furthermore, they lacked the ability to recognize emotions in real time and dynamically adjust commentary based on those emotions, making it difficult to provide an optimal viewing experience for each user.

[0332] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game video in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for transmitting the generated commentary to a user terminal, means for recognizing and analyzing the user's emotional state, and means for dynamically adjusting the commentary content based on the user's emotional state. This allows the user to receive detailed real-time commentary tailored to a specific player, and further provides a more satisfying viewing experience through commentary customized to the user's emotions.

[0333] 1. "Game Footage" means real-time video data including the plays and player actions during a game.

[0334] 2. "Analyzing" means automatically analyzing received game video data using a computer model to extract specific information.

[0335] 3. "Specific target player" refers to an individual player who is of interest to the user and for whom the system will specialize in analyzing and providing information.

[0336] 4. "Extracting information" refers to extracting important movement and positional information related to a specific player from the real-time analyzed data.

[0337] 5. "Generating commentary" means automatically generating a commentary in text or audio format based on the extracted information.

[0338] 6. "User Terminal" means an electronic device capable of receiving and displaying data and information from the System, including a smartphone or computer.

[0339] 7. "Emotional state" refers to the user's psychological and emotional state as perceived through facial expressions, tone of voice, etc.

[0340] 8. "Recognize and analyze" means detecting the user's emotional state in real time and analyzing that information.

[0341] 9. "Dynamic adjustment of commentary content" means automatically changing the content or tone of commentary in response to the user's perceived emotional state.

[0342] This invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system receives game footage in real time, analyzes it using a generative AI model, extracts information about specific athletes, and generates commentary from that information. The generated commentary is then sent to the user's device and dynamically adjusted according to the user's emotional state.

[0343] The server receives game footage in real time and analyzes it frame by frame. It uses a generative AI model to detect the movements and location information of specific target players. For example, in a soccer game, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0344] From the analysis results, the server extracts information about the specific target player. In this process, the specific target player's important actions such as dribbling, shooting, passing, etc., as well as their position information, are recorded.

[0345] The server generates a commentary in natural language based on the extracted information. For example, it might say, "Player A is dribbling forward, dodging the defender!" This commentary is also converted into audio data using speech synthesis technology.

[0346] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays it back as audio. This allows the user to keep track of the latest status of a specific target player in real time during a match. For example, a smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the commentary back as audio.

[0347] The user device is equipped with an emotion engine that recognizes the user's emotional state in real time. The emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[0348] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to be even more exciting. Commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!" This allows users to receive a more satisfying and individually customized commentary.

[0349] For example, the following prompt might be generated:

[0350] "Player A is dribbling forward, dodging the defender!" "Player A is closing in on the goal!"

[0351] This system allows users to receive detailed action of specific players in real time and commentary based on their own emotions, dramatically improving the sports viewing experience.

[0352] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0353] Step 1:

[0354] The user starts the application and selects a specific target player. The user then selects their favorite player from the displayed player list and sends that selection to the server. The input is the player list and the user's selection, and the output is the selected player information being sent to the server.

[0355] Step 2:

[0356] The server receives live-streamed game footage in real time. The input is the game video data, and the output is each frame of that video data. The server splits the game video into frames and prepares them for analysis.

[0357] Step 3:

[0358] The server analyzes each frame of the received game video using a generative AI model to detect the movements of specific target players. The input is video data for each frame, and the output is the position information and action data of the specific target player. The server uses the generative AI model to identify important actions such as dribbling, shooting, and passing.

[0359] Step 4:

[0360] The server extracts information related to the specific target player from the analysis results. The input is the analyzed frame data, and the output is the important actions and position information of the specific target player. The server organizes this information as structured data.

[0361] Step 5:

[0362] The server generates natural language commentary based on the extracted information. The input is the key actions and positional information of a specific target player, and the output is commentary text. The server uses a natural language generation model to create commentary such as, "Player A is dribbling forward, dodging the defender!"

[0363] Step 6:

[0364] The server converts the generated commentary text into audio data using speech synthesis technology. The input is the commentary text and the output is audio data. The server then generates the audio file and prepares it to be sent to the user.

[0365] Step 7:

[0366] The server sends the generated commentary (text and audio data) to the user terminal. The input is the commentary text and audio data, and the output is the commentary data received by the user terminal. The user terminal displays the received commentary in real time and starts playing the audio.

[0367] Step 8:

[0368] The user terminal uses an emotion engine to recognize the user's emotional state in real time. The input is the user's facial expression data and voice tone, and the output is the recognized emotional state data. The emotion engine analyzes the user's facial expression and voice tone to identify the emotional state.

[0369] Step 9:

[0370] The server receives the user's emotional state and dynamically adjusts the commentary content based on that information. The input is emotional state data, and the output is adjusted commentary text and audio data. For example, if the user is excited, the commentary content will be changed to something like, "Player A is showing off some amazing dribbling!"

[0371] Step 10:

[0372] The server then sends the adjusted commentary back to the user's device, where the user receives the customized commentary in real time. The input is the adjusted commentary data, and the output is the adjusted commentary information displayed on the user's device, allowing the user to enjoy a highly personalized viewing experience.

[0373] 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.

[0374] 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.

[0375] 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.

[0376] [Second embodiment]

[0377] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0378] 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.

[0379] 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).

[0380] 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.

[0381] 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.

[0382] 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).

[0383] 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.

[0384] 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.

[0385] 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.

[0386] 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.

[0387] 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.

[0388] 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."

[0389] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete). Below, we will explain how the program of this system works, using concrete examples.

[0390] Setting favorite players on user devices

[0391] To use the system, a user first selects a specific target player to support. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the application on their smartphone and selects "Player A" from the player list.

[0392] The user terminal acquires this selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0393] Real-time analysis of game footage by server

[0394] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, it detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0395] Identifying information about a specific target player by the server

[0396] The server extracts information related to the specific target player from the analysis results. In this process, the server precisely records the specific target player's important actions (e.g., dribbling, shooting, passing), their position information, and the surrounding situation. This allows users to obtain the information they want in real time.

[0397] Server-based commentary generation

[0398] Based on the extracted information, the server generates a commentary in natural language. The generated commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0399] Transmission of commentary generated by the server and display on the user's device

[0400] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to check the latest status of their favorite player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0401] This system allows users to obtain timely information specific to their favorite athletes, eliminating the frustrations of conventional live sports broadcasts. In this way, the present invention significantly improves the user's sports viewing experience.

[0402] The processing flow will be explained below.

[0403] Step 1:

[0404] The user launches the application and selects a specific target player. The user selects their favorite target player from the player list in the application and enters the selection information into the device.

[0405] Step 2:

[0406] The terminal sends the selection information to the server. The terminal generates a request including the identification information (e.g., player ID, name) of the specific target player selected by the user and sends it to the server.

[0407] Step 3:

[0408] The server receives live-streamed game footage in real time. The server continuously receives the game footage data stream and stores it in a buffer.

[0409] Step 4:

[0410] The server analyzes the game footage frame by frame, and the server's generative AI model processes the received footage frame by frame to identify the positions and movements of players in each frame.

[0411] Step 5:

[0412] The server extracts information about the specific target player. The server extracts information related to the specific target player selected by the user (e.g., position data, ball possession status, actions) from the analyzed video data.

[0413] Step 6:

[0414] The server generates a commentary based on the extracted information. The server generates a commentary in natural language using the information on the specific target player, and then converts the commentary into audio data using speech synthesis technology.

[0415] Step 7:

[0416] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0417] Step 8:

[0418] The terminal displays and plays the commentary received. The terminal displays the text commentary received from the server on the screen and plays the audio commentary, allowing the user to understand the real-time situation of a specific target player.

[0419] Example 1

[0420] 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."

[0421] Conventional sports viewing systems have difficulty obtaining real-time information about specific athletes that users are interested in. Furthermore, conventional systems lack sufficient video analysis and commentary generation for specific athletes selected by users, resulting in an unsatisfactory sports viewing experience for users. Furthermore, the mechanisms for providing generated information to user devices in real time are also inadequate.

[0422] 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.

[0423] In this invention, the server includes means for receiving and analyzing game footage in real time, means for identifying specific target players based on the analysis results, means for extracting information about the identified specific target players, means for generating a commentary in natural language based on the extracted information, means for converting the generated natural language commentary into audio data, and means for transmitting the generated commentary to a user terminal, thereby enabling users to obtain information about specific players in real time and significantly improving their viewing experience.

[0424] "Game video" refers to video data that records the state of a sports game.

[0425] "Means for receiving and analyzing in real time" refers to a combination of hardware and software for receiving game footage in real time via a network and sequentially analyzing the footage.

[0426] The "means for identifying a specific target player based on the analysis results" refers to an algorithm and program for identifying a specific player from the analyzed game video data.

[0427] The "means for extracting information about the identified specific target player" is a program for extracting related information such as the position information and movements of the specific player, and their positional relationship with other players, from the analysis results.

[0428] The "means for generating commentary in natural language" refers to an algorithm and program for generating commentary in text format based on the extracted information.

[0429] The "means for converting the commentary in natural language into voice data" is a voice synthesis technology for converting the generated commentary in text format into voice data.

[0430] "Means for transmitting to user terminal" refers to network technologies and protocols for transmitting the generated commentary to the user's device in real time.

[0431] The "means for the user to select a specific target player" refers to an interface for the user to select a favorite player and a program for confirming the selection.

[0432] The "means for transmitting selected information to the server" refers to a network technology for transmitting information about the specific target player selected by the user to the server.

[0433] The "means for generating a request including identification information of a selected specific target player" is a program for generating a data request including identification information about the selected player.

[0434] The "means for transmitting the generated request to the server" refers to a network technology for transmitting the generated request to the server.

[0435] The "means for receiving and displaying the live commentary from the server" refers to a program and interface for receiving the live commentary sent from the server at the user terminal and displaying it on the screen.

[0436] "Means for generating information in natural language using a generative AI model" means a program that uses a generative AI model (e.g., GPT, BERT) to generate commentary in natural language based on extracted information.

[0437] The present invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system combines a user terminal, a server, and a generative AI model to analyze game footage in real time and provide information.

[0438] Users access the system through a smartphone application. When the application is launched, the user can access an interface for selecting a favorite player. When the user selects a specific player (e.g., "Player A") from the player list and confirms the selection, the user's device acquires this information. The device generates a request including the identification information (player ID) of the selected specific target player and sends it to the server. This request is sent using the HTTP POST method.

[0439] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API) and uses a GPU to perform high-speed image analysis. The software used includes generative AI models (e.g., YOLO and OpenPose). Using these models, the server analyzes the game footage frame by frame to detect the positions and movements of specific target players. For example, in a soccer game, the analysis identifies the position and actions (e.g., dribbling, shooting, passing) of "Player A."

[0440] The server extracts information related to "Player A" from the analysis results. During this extraction process, "Player A's" location information, actions, and surrounding circumstances are recorded in detail. This information is used to grasp the data the user wants to know in real time.

[0441] The server generates a commentary in natural language based on the extracted information. A generative AI model (e.g., GPT-3 or BERT) is used for natural language generation. By inputting a prompt sentence such as "Player A is dribbling," the generative AI model generates specific commentary text such as "Player A is dribbling forward, dodging the defender!" This text commentary is then converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly).

[0442] The generated commentary (text and audio data) is sent from the server to the user's device. WebSocket and HTTP / 2 protocols are used to transmit data in real time. The user's device immediately processes the received commentary data. The received commentary text is displayed on the display, and the audio commentary is played through the speaker.

[0443] As a concrete example, consider the case where a user selects a soccer player, "Player A," as their favorite player. The user selects "Player A" using a smartphone app, and that information is sent to the server. The server receives the game footage, analyzes it using a generative AI model, and identifies the scene where "Player A" is dribbling in front of the goal. As a result, a commentary such as "Player A is approaching the goal!" is generated and immediately sent to the user's smartphone. The user's smartphone displays this commentary and simultaneously plays it back as audio.

[0444] Examples of prompts:

[0445] "Player A starts dribbling at 15:42"

[0446] "Player A is dribbling forward past the defender!"

[0447] As described above, the system of the present invention allows users to obtain real-time information specific to a particular athlete, greatly enhancing the sports viewing experience.

[0448] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0449] Step 1:

[0450] User-defined favorite players

[0451] The user launches the smartphone application and sets their favorite player. Specifically, they select a specific target player (for example, "Player A") from the player list in the application and confirm the selection. The input data is the player list and player ID, and the output data is the identification information (player ID) of the selected player. The user selects "Player A," and that ID is stored within the application.

[0452] Step 2:

[0453] Sending a request from the device to the server

[0454] The terminal generates a request including the identification information (player ID) of the specific target player selected by the user. This request is sent to the server using the HTTP POST method. The input data is the player ID, and the output data is the request sent to the server. The terminal converts the player ID into a JSON object and sends it to the URL https: / / api.example.com / select-player.

[0455] Step 3:

[0456] Real-time analysis of game footage by server

[0457] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API). It uses a GPU to process the video data frame by frame and analyzes player movements using a generative AI model (e.g., YOLO or OpenPose). The input data is the live stream video, and the output data is player position and movement information. The server analyzes the game footage frame by frame to identify the position and movement of "Player A."

[0458] Step 4:

[0459] Server extracts information about specific target players

[0460] The server extracts information related to a specific target player ("Player A") from the analysis results. The input data is the frame analysis results, and the output data is "Player A's" position information, actions, and surrounding circumstances. The server records this information in a database, allowing users to grasp the information they want to know in real time.

[0461] Step 5:

[0462] Server-based commentary generation

[0463] The server generates commentary in natural language based on the extracted information. Using a generative AI model (e.g., GPT-3 or BERT), a prompt sentence is input and specific commentary text is generated. The input data is the extracted player information, and the output data is commentary text in natural language. For example, the server generates commentary text such as, "Player A is dribbling forward, dodging the defender!"

[0464] Step 6:

[0465] Live audio conversion by the server

[0466] The generated commentary text is converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The input data is the commentary text, and the output data is audio data. The server converts the generated commentary text into an audio file.

[0467] Step 7:

[0468] Live data transmission from the server to the device

[0469] The server transmits the generated commentary text and audio data to the user terminal in real time. The input data is the commentary text and audio data, and the output data is the transmitted commentary data. This transmission is performed using the WebSocket or HTTP / 2 protocol.

[0470] Step 8:

[0471] Live display and audio playback on the device

[0472] The device processes the commentary data received from the server, displays the text on the display, and plays the audio from the speaker. The input data is the commentary text and audio data, and the output data is the displayed text and the played audio. For example, the commentary text "Player A is closing in on the goal!" is displayed on the screen, and the content is played back as audio.

[0473] (Application example 1)

[0474] 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."

[0475] In conventional sports commentary, it was difficult to obtain detailed information about specific players in real time. In addition, users were unable to receive timely information specific to their favorite players, and were only provided with general information that was in line with the progress of the entire game. This limited users' sports viewing experience, preventing them from enjoying the excitement of focusing on the play of their favorite players in real time.

[0476] 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.

[0477] In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about specific target players based on the analysis results, means for generating commentary about the specific target players based on the extracted information, means for transmitting the generated commentary to a user terminal, and means for displaying and playing back the received commentary on the user terminal as audio. This allows users to receive detailed information about the play of their favorite players in real time, providing a more exciting sports viewing experience through both their eyes and their ears.

[0478] "Game footage" is video data captured during a sporting event.

[0479] "Receiving in real time" means obtaining data without delay while the match is in progress.

[0480] "Means for analysis" refers to a device or program that can extract the movements and positional information of specific players from the received game footage.

[0481] A "specific target athlete" is a specific athlete designated by the user.

[0482] "Means for extracting information" refers to devices or programs that extract data related to specific target players from the analysis results of game footage.

[0483] "Means for generating commentary" refers to a device or program that creates commentary in natural language from the extracted player information.

[0484] A "user terminal" is an electronic device used by a user, and examples include smartphones and tablets.

[0485] The "transmitting means" refers to a communication device or program for delivering the generated commentary to the user terminal.

[0486] "Means for displaying and playing audio" refers to a device or program that displays the received commentary on a screen and plays the audio through speakers or headphones.

[0487] "User-selected means" refers to the interface or operation method that allows the user to specify their favorite player.

[0488] "Means for sending to server" refers to a communication device or program for sending the information selected by the user to the central server.

[0489] "Natural language commentary" is commentary that expresses information about a specific player in natural language that is easy for people to understand.

[0490] "Means for converting into audio data" refers to a device or program that converts the generated natural language text into an audio file.

[0491] This invention is a system that provides real-time commentary specific to a specific athlete. This system provides users with detailed information such as the movements and positions of the specific athlete in real time, thereby improving the sports viewing experience. Specific embodiments of this system are described below.

[0492] System Configuration

[0493] The server includes the following means:

[0494] 1. Means of receiving and analyzing game footage in real time:

[0495] The server receives live-streamed game footage in real time and uses deep learning models such as TensorFlow for video analysis.

[0496] 2. Methods for extracting information on specific target players based on the analysis results:

[0497] The received video is analyzed frame by frame to extract the movement and position information of specific target players, which includes player identification and action recognition.

[0498] 3. A means for generating commentary about a specific target player based on the extracted information:

[0499] Based on the extracted information, a live commentary text is generated using natural language processing technology, using generative AI models such as Transformers for natural language generation.

[0500] 4. Means for transmitting the generated commentary to the user terminal:

[0501] The generated text and voice data are sent to the user's device using real-time data transfer technology (e.g., WebSocket) over the Internet.

[0502] The user terminal includes the following means:

[0503] 1. Means for displaying and playing back received commentary:

[0504] The user device (e.g., a smartphone) displays the received commentary text on its screen and plays the audio data through speakers or headphones using a voice playback and speech synthesis library such as pygame or gTTS.

[0505] 2. How users select specific target players:

[0506] The user selects their favorite player through the application, and the selection information is sent to the server, where the player is set as a specific target player.

[0507] Specific examples

[0508] Specific examples are given below.

[0509] Let's take the example of a scenario where a user is using "Favorite Player Live Assist" on their smartphone to watch a soccer match. The user selects Player A from the application's list and sets him as the specific target player. Meanwhile, the server analyzes the game footage in real time and identifies Player A's movements. For example, when it captures a scene where Player A has the ball and begins to dribble, it generates a commentary such as, "Player A is dribbling forward, dodging the defender." This commentary is immediately sent to the smartphone and displayed on the screen, while also being played back as audio using speech synthesis technology.

[0510] Example prompts for generative AI models

[0511] "Player A is dribbling forward past the defender. Please provide a commentary of the game situation."

[0512] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0513] Step 1:

[0514] The user selects a specific target player

[0515] The user launches the application and selects a specific target player from a list. The user selects their favorite player from the player list (loaded from the database) by tapping on it. As input, the player's identification information is generated and sent to the server. As output, the player ID is selected and sent to the server.

[0516] Step 2:

[0517] Send the selected information to the server

[0518] The terminal sends the identification information of the specific target player selected by the user to the server. The server receives the information and sets the specified player as the specific target player. The player ID is sent from the terminal as input, and the player information is saved in the database in the server as output.

[0519] Step 3:

[0520] The server receives and analyzes the match footage in real time.

[0521] The server receives live-streamed game footage in real time and analyzes it using a deep learning model (such as TensorFlow). Live video data is given as input, and frame-by-frame analysis results are output. Specifically, player movements and position information are identified for each frame.

[0522] Step 4:

[0523] Extract information on specific target players based on the analysis results

[0524] The server extracts information related to a specific target player from the analysis results. The analyzed data of all players is given as input, and data showing the actions and positions of the specific target player is obtained as output. Specifically, the player's actions (e.g., dribbling, passing, shooting) and current position information are extracted.

[0525] Step 5:

[0526] Generate commentary based on the extracted information

[0527] Based on the extracted information, the server uses a generative AI model (e.g., GPT-3) to generate commentary text in natural language. The input is given as movement data of a specific target player, and the output is generated commentary text. Specifically, the generative AI creates text using the prompt sentence, "Player A is dribbling forward, dodging the defender. Please provide commentary on the situation in the game."

[0528] Step 6:

[0529] Send the generated commentary to the user's device

[0530] The server transmits the generated commentary text and audio data to the user terminal. The commentary text and audio files are stored in the server as input, and are transmitted to the user terminal as output. Specifically, the data is transmitted in real time via the Internet.

[0531] Step 7:

[0532] The user device displays the commentary and plays it back with audio.

[0533] The user device displays the received commentary text on the screen and plays the audio data through speakers or headphones. The input is the received text and audio file, and the output is the display of text on the screen and audio playback. The user can follow the movements of a specific player in real time through both sight and sound.

[0534] 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.

[0535] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete), and further combines it with an emotion engine that recognizes the user's emotions. How the program of this system works will be explained using concrete examples.

[0536] Setting favorite players on user devices

[0537] To use the system, a user must first select a specific target player. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the smartphone app and selects "Player A" from the player list.

[0538] The user terminal acquires the selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0539] Real-time analysis of game footage by server

[0540] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0541] Identifying information about a specific target player by the server

[0542] The server extracts information related to the specific target player from the analysis results. During this process, the key actions of the specific target player (e.g., dribbling, shooting, passing), their position, and the surrounding situation are precisely recorded. This allows users to obtain the information they want in real time.

[0543] Server-based commentary generation

[0544] Based on the extracted information, the server generates commentary in natural language. This commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0545] Transmission of commentary generated by the server and display on the user's device

[0546] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to keep track of the latest status of a specific target player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the same commentary in audio.

[0547] Recognizing user emotions with an emotion engine

[0548] The user device is equipped with an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the user's facial expressions and voice tone in real time to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[0549] Adjusting commentary based on emotion information

[0550] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to make it even more exciting. A rich commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!"

[0551] This system allows users to understand the status of specific athletes in real time and receive commentary that is customized to their individual emotions, eliminating frustrations with traditional sports commentary and significantly improving the user's sports viewing experience.

[0552] The processing flow will be explained below.

[0553] Step 1:

[0554] The user launches the application and selects a specific player. The user selects "Player A" from the player list in the app.

[0555] Step 2:

[0556] The terminal transmits the selection information to the server, and the terminal generates a request including the identification information of the specific target player and transmits it to the server.

[0557] Step 3:

[0558] The server receives live-streamed game footage in real time and stores the game footage data stream in a buffer.

[0559] Step 4:

[0560] The server analyzes the game footage frame by frame, and the server's generative AI model processes each frame to identify the players' positions and movements.

[0561] Step 5:

[0562] The server extracts information about the specific target player, and extracts information related to the specific target player from the analysis results (e.g., position data, ball possession status, actions).

[0563] Step 6:

[0564] The server generates commentary based on the extracted information, using natural language generation technology to generate commentary such as, "Player A is dribbling forward, dodging the defender!"

[0565] Step 7:

[0566] The server converts the generated commentary into audio data, and the generated text commentary is converted into audio data using voice synthesis technology.

[0567] Step 8:

[0568] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0569] Step 9:

[0570] The device displays and plays the commentary it receives. The device displays the received text commentary on the screen and plays the received audio commentary.

[0571] Step 10:

[0572] The device's emotion engine recognizes the user's emotions by analyzing the user's facial expressions and tone of voice to identify the user's emotional state.

[0573] Step 11:

[0574] The device transmits emotion information to the server. The emotion engine transmits the user's emotion information identified by the device to the server.

[0575] Step 12:

[0576] The server adjusts the commentary based on the emotional information. The server receives the user's emotional state and dynamically adjusts the commentary content based on the information.

[0577] Step 13:

[0578] The server transmits the adjusted commentary to the user terminal, and the server transmits the adjusted commentary data to the user terminal again.

[0579] Step 14:

[0580] The terminal displays and plays the adjusted commentary. The terminal displays the received adjusted commentary on the screen and plays it back as audio.

[0581] Example 2

[0582] 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."

[0583] Conventional sports commentary systems make it difficult for users to receive real-time information about specific athletes, and they are unable to provide customized commentary that takes into account the user's emotional state. This results in a standardized viewing experience for users, and makes it difficult to meet individual needs.

[0584] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for recognizing the user's emotional state, means for dynamically adjusting the commentary content based on the recognized emotional information, and means for transmitting the generated commentary to the user terminal. This not only enables the user to receive detailed real-time information about a specific player, but also makes it possible to provide a commentary customized according to the user's emotions.

[0585] "Game footage" refers to video data that records the progress of a sports game in real time.

[0586] The "analysis means" refers to a technology or device that analyzes the input game video and extracts important information from each frame.

[0587] "Specific target athlete" refers to a specific athlete selected by the user.

[0588] "Means for extracting information" refers to a technique or method for extracting data (position information, movements, etc.) related to a specific target player from the analysis results.

[0589] The "means for generating commentary" is a technique or method for describing the actions of a specific target player and the situation of the game in natural language based on the extracted data.

[0590] "User's emotional state" refers to the emotional state obtained by analyzing the user's facial expression, voice tone, and other biometric information.

[0591] The "means for recognizing emotional states" refers to a technology or device that analyzes the user's facial expressions and tone of voice to identify emotions.

[0592] A "means for dynamically adjusting commentary content" is a technique or method for changing commentary content in real time based on recognized emotion information.

[0593] "User terminal" refers to a computer, smartphone, tablet, or other mobile device used by a User.

[0594] The "generated commentary" is text and audio data of a narration generated based on the progress of the match.

[0595] The "transmitting means" refers to a technique or method for transmitting the generated commentary to a user terminal.

[0596] This invention is a system that allows users to receive real-time commentary on specific athletes (hereinafter referred to as "specific target athletes") and further customizes the commentary content by recognizing the user's emotions. The program processing of this system is explained in detail below.

[0597] To use the system, a user must first select a specific target player. The user launches the smartphone app and selects a specific target player from the player list. For example, the user selects "Player A." At this time, the smartphone generates a request including the player's identification information and sends it to the server.

[0598] The server receives this request and uses a generative AI model such as YOLOv5 to analyze each frame of the live-streamed game footage in real time, identifying the position and movement of specific target players and extracting important actions (e.g., dribbling, shooting, passing).

[0599] Based on the extracted information, the server generates commentary for the specific target player using a natural language generation model such as GPT-3. For example, a commentary such as "Player A is dribbling forward, dodging the defender!" is generated. This commentary is then converted into audio data using voice synthesis technology such as Google Cloud Text-to-Speech.

[0600] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays this commentary in real time and plays the audio. For example, the user's smartphone might display the text "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0601] Furthermore, the user device is equipped with an emotion engine that analyzes the user's facial expressions and voice tone in real time to identify their emotional state. The emotion engine uses TensorFlow and other tools to recognize the user's surprise, excitement, joy, etc., and sends this information to the server. The server receives the user's emotional information and, for example, if the user is very excited, adjusts the commentary to make it more exciting. Commentary that corresponds to the user's emotional state is generated, such as "Player A is showing off some amazing dribbling!"

[0602] This system not only allows users to keep up with the latest game progress of specific players in real time, but also allows them to receive commentary that is customized to their emotions at the time. This technology meets individual needs that traditional sports commentary could not meet, significantly improving the user's viewing experience.

[0603] Examples of specific examples and prompts

[0604] For example, if a user "Tanaka" has set a specific target player "Player A" on his smartphone and is watching a soccer game, and Tanaka looks surprised when "Player A" shoots the ball:

[0605] 1. The user device detects Tanaka's surprised expression, and the emotion engine analyzes it.

[0606] 2. The emotion engine sends the recognized emotion information to the server.

[0607] 3. The server adjusts the commentary based on this information, generating a commentary such as "Player A took a great shot!"

[0608] 4. This live commentary will be displayed in real time on Tanaka's smartphone and played back as audio.

[0609] Example prompt sentence:

[0610] "Generate a commentary of a game played by Player A. If the user seems excited, make the commentary even more exciting."

[0611] The above is a specific embodiment of the present invention, which allows users to enjoy detailed game situations of specific target players and commentary customized to their own emotions in real time.

[0612] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0613] Program processing flow

[0614] Step 1: User's Favourite Player

[0615] 1. The user launches the smartphone app and selects a specific target player from the player list.

[0616] Input: User selects a player within the app

[0617] Output: Identification information of the specific target player is generated.

[0618] Specific operation: When the user selects "Player A" on the app screen, the device generates the identification information of the specific target player (e.g., { "player_id": "A123"}).

[0619] 2. The terminal generates a request including the selection information and sends it to the server.

[0620] Input: Identification information of specific target player

[0621] Output: The request sent to the server

[0622] Specific operation: The device sends a request containing the device-generated identification information to the server in the following format: { "action": "select_player", "player_id": "A123"}

[0623] Step 2: Real-time analysis of game footage by the server

[0624] 1. The server receives live footage of the match.

[0625] Input: Live video stream

[0626] Output: Image data for each frame

[0627] Specific operation: The server uses tools such as FFmpeg to divide the live video into frames and generate image data for each frame.

[0628] 2. The server analyzes each frame using a generative AI model to detect the position and movement of specific target players.

[0629] Input: Frame-by-frame image data

[0630] Output: Position and action information of specific target players

[0631] Specific behavior: Using a generative AI model such as YOLOv5, each frame is analyzed to detect the position (e.g., coordinate information) and actions (e.g., dribbling, shooting) of specific target players.

[0632] Step 3: The server identifies information about the specific target player.

[0633] 1. The server extracts important actions of specific target players from the analysis results.

[0634] Input: Detected location and action information

[0635] Output: List of important actions of a specific target player

[0636] Specific actions: Extract the main actions of a specific target player from the analysis results (e.g., { "time": "10:12", "action": "dribble"}) and save them in list format.

[0637] Step 4: Server-generated commentary

[0638] 1. The server generates commentary text based on the extracted information.

[0639] Input: List of important actions for a specific target player

[0640] Output: Commentary text

[0641] Specific Actions: Use generative AI models such as GPT-3 to generate commentary text based on specific player actions and game situations (e.g., "Player A is dribbling forward, dodging the defender!").

[0642] 2. The server converts the generated text into audio data.

[0643] Input: Play-by-play text

[0644] Output: Audio data

[0645] Specific operation: The generated commentary text is converted into audio data using Google Cloud Text-to-Speech or similar.

[0646] Step 5: Transmission of the generated commentary by the server and display on the user's device

[0647] 1. The server sends the generated commentary (text and audio data) to the user terminal.

[0648] Input: Live text and audio data

[0649] Output: Live data sent to the user's device

[0650] Specific operation: A network communication protocol (e.g., HTTP, WebSocket) is used to send live data to the user terminal.

[0651] 2. The device displays the received commentary in real time and plays it back as audio.

[0652] Input: Live data from the server

[0653] Output: Running text displayed on the screen and audio played back

[0654] Specific operation: The received commentary text is displayed on the device screen and the audio data is played back.

[0655] Step 6: Recognizing user emotions with the emotion engine

[0656] 1. The device collects the user's facial expressions and voice tone in real time.

[0657] Input: User's facial and voice data

[0658] Output: Collected real-time emotion data

[0659] Specific operation: The device's camera and microphone are used to collect the user's facial expressions and voice tone in real time.

[0660] 2. The device analyzes the collected emotional data and identifies the user's emotional state.

[0661] Input: Collected emotion data

[0662] Output: Parsed emotional state information

[0663] Specific operation: Analyze the user's emotional state (e.g., surprise, excitement) using an emotion recognition engine such as TensorFlow.

[0664] 3. The device sends the analyzed emotion information to the server.

[0665] Input: Parsed emotional state information

[0666] Output: Emotion information sent to the server

[0667] Specific operation: A network communication protocol is used to transmit emotional state information to the server.

[0668] Step 7: Adjusting commentary based on emotional information

[0669] 1. The server receives the user's emotional information and adjusts the commentary content based on that emotional information.

[0670] Input: User's emotional information

[0671] Output: Adjusted commentary text

[0672] Specific operation: Based on the user's emotional information, the commentary text is regenerated using a generative AI model such as GPT-3 and the content is adjusted.

[0673] 2. The server sends the adjusted commentary back to the user's device.

[0674] Input: Adjusted play-by-play text

[0675] Output: Adjusted live data sent to user device

[0676] Specific operation: The adjusted live data is sent back to the user's terminal, and the terminal displays and plays it.

[0677] The above are the specific processing steps of the system program, and this series of processes enables users to receive real-time information on specific players and customized commentary according to their emotions.

[0678] (Application example 2)

[0679] 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."

[0680] Conventional sports commentary systems lacked commentary tailored to specific athletes, making it difficult to provide information tailored to users' interests. Furthermore, they lacked commentary tailored to the user's emotions, resulting in a limited viewing experience and low satisfaction. Furthermore, they lacked the ability to recognize emotions in real time and dynamically adjust commentary based on those emotions, making it difficult to provide an optimal viewing experience for each user.

[0681] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game video in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for transmitting the generated commentary to a user terminal, means for recognizing and analyzing the user's emotional state, and means for dynamically adjusting the commentary content based on the user's emotional state. This allows the user to receive detailed real-time commentary tailored to a specific player, and further provides a more satisfying viewing experience through commentary customized to the user's emotions.

[0682] 1. "Game Footage" means real-time video data including the plays and player actions during a game.

[0683] 2. "Analyzing" means automatically analyzing received game video data using a computer model to extract specific information.

[0684] 3. "Specific target player" refers to an individual player who is of interest to the user and for whom the system will specialize in analyzing and providing information.

[0685] 4. "Extracting information" refers to extracting important movement and positional information related to a specific player from the real-time analyzed data.

[0686] 5. "Generating commentary" means automatically generating a commentary in text or audio format based on the extracted information.

[0687] 6. "User Terminal" means an electronic device capable of receiving and displaying data and information from the System, including a smartphone or computer.

[0688] 7. "Emotional state" refers to the user's psychological and emotional state as perceived through facial expressions, tone of voice, etc.

[0689] 8. "Recognize and analyze" means detecting the user's emotional state in real time and analyzing that information.

[0690] 9. "Dynamic adjustment of commentary content" means automatically changing the content or tone of commentary in response to the user's perceived emotional state.

[0691] This invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system receives game footage in real time, analyzes it using a generative AI model, extracts information about specific athletes, and generates commentary from that information. The generated commentary is then sent to the user's device and dynamically adjusted according to the user's emotional state.

[0692] The server receives game footage in real time and analyzes it frame by frame. It uses a generative AI model to detect the movements and location information of specific target players. For example, in a soccer game, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0693] From the analysis results, the server extracts information about the specific target player. In this process, the specific target player's important actions such as dribbling, shooting, passing, etc., as well as their position information, are recorded.

[0694] The server generates a commentary in natural language based on the extracted information. For example, it might say, "Player A is dribbling forward, dodging the defender!" This commentary is also converted into audio data using speech synthesis technology.

[0695] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays it back as audio. This allows the user to keep track of the latest status of a specific target player in real time during a match. For example, a smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the commentary back as audio.

[0696] The user device is equipped with an emotion engine that recognizes the user's emotional state in real time. The emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[0697] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to be even more exciting. Commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!" This allows users to receive a more satisfying and individually customized commentary.

[0698] For example, the following prompt might be generated:

[0699] "Player A is dribbling forward, dodging the defender!" "Player A is closing in on the goal!"

[0700] This system allows users to receive detailed action of specific players in real time and commentary based on their own emotions, dramatically improving the sports viewing experience.

[0701] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0702] Step 1:

[0703] The user starts the application and selects a specific target player. The user then selects their favorite player from the displayed player list and sends that selection to the server. The input is the player list and the user's selection, and the output is the selected player information being sent to the server.

[0704] Step 2:

[0705] The server receives live-streamed game footage in real time. The input is the game video data, and the output is each frame of that video data. The server splits the game video into frames and prepares them for analysis.

[0706] Step 3:

[0707] The server analyzes each frame of the received game video using a generative AI model to detect the movements of specific target players. The input is video data for each frame, and the output is the position information and action data of the specific target player. The server uses the generative AI model to identify important actions such as dribbling, shooting, and passing.

[0708] Step 4:

[0709] The server extracts information related to the specific target player from the analysis results. The input is the analyzed frame data, and the output is the important actions and position information of the specific target player. The server organizes this information as structured data.

[0710] Step 5:

[0711] The server generates natural language commentary based on the extracted information. The input is the key actions and positional information of a specific target player, and the output is commentary text. The server uses a natural language generation model to create commentary such as, "Player A is dribbling forward, dodging the defender!"

[0712] Step 6:

[0713] The server converts the generated commentary text into audio data using speech synthesis technology. The input is the commentary text and the output is audio data. The server then generates the audio file and prepares it to be sent to the user.

[0714] Step 7:

[0715] The server sends the generated commentary (text and audio data) to the user terminal. The input is the commentary text and audio data, and the output is the commentary data received by the user terminal. The user terminal displays the received commentary in real time and starts playing the audio.

[0716] Step 8:

[0717] The user terminal uses an emotion engine to recognize the user's emotional state in real time. The input is the user's facial expression data and voice tone, and the output is the recognized emotional state data. The emotion engine analyzes the user's facial expression and voice tone to identify the emotional state.

[0718] Step 9:

[0719] The server receives the user's emotional state and dynamically adjusts the commentary content based on that information. The input is emotional state data, and the output is adjusted commentary text and audio data. For example, if the user is excited, the commentary content will be changed to something like, "Player A is showing off some amazing dribbling!"

[0720] Step 10:

[0721] The server then sends the adjusted commentary back to the user's device, where the user receives the customized commentary in real time. The input is the adjusted commentary data, and the output is the adjusted commentary information displayed on the user's device, allowing the user to enjoy a highly personalized viewing experience.

[0722] 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.

[0723] 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.

[0724] 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.

[0725] [Third embodiment]

[0726] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0727] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0728] 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).

[0729] 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.

[0730] 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.

[0731] 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).

[0732] 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.

[0733] 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.

[0734] 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.

[0735] 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.

[0736] 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.

[0737] 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."

[0738] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete). Below, we will explain how the program of this system works, using concrete examples.

[0739] Setting favorite players on user devices

[0740] To use the system, a user first selects a specific target player to support. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the application on their smartphone and selects "Player A" from the player list.

[0741] The user terminal acquires this selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0742] Real-time analysis of game footage by server

[0743] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, it detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0744] Identifying information about a specific target player by the server

[0745] The server extracts information related to the specific target player from the analysis results. In this process, the server precisely records the specific target player's important actions (e.g., dribbling, shooting, passing), their position information, and the surrounding situation. This allows users to obtain the information they want in real time.

[0746] Server-based commentary generation

[0747] Based on the extracted information, the server generates a commentary in natural language. The generated commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0748] Transmission of commentary generated by the server and display on the user's device

[0749] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to check the latest status of their favorite player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0750] This system allows users to obtain timely information specific to their favorite athletes, eliminating the frustrations of conventional live sports broadcasts. In this way, the present invention significantly improves the user's sports viewing experience.

[0751] The processing flow will be explained below.

[0752] Step 1:

[0753] The user launches the application and selects a specific target player. The user selects their favorite target player from the player list in the application and enters the selection information into the device.

[0754] Step 2:

[0755] The terminal sends the selection information to the server. The terminal generates a request including the identification information (e.g., player ID, name) of the specific target player selected by the user and sends it to the server.

[0756] Step 3:

[0757] The server receives live-streamed game footage in real time. The server continuously receives the game footage data stream and stores it in a buffer.

[0758] Step 4:

[0759] The server analyzes the game footage frame by frame, and the server's generative AI model processes the received footage frame by frame to identify the positions and movements of players in each frame.

[0760] Step 5:

[0761] The server extracts information about the specific target player. The server extracts information related to the specific target player selected by the user (e.g., position data, ball possession status, actions) from the analyzed video data.

[0762] Step 6:

[0763] The server generates a commentary based on the extracted information. The server generates a commentary in natural language using the information on the specific target player, and then converts the commentary into audio data using speech synthesis technology.

[0764] Step 7:

[0765] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0766] Step 8:

[0767] The terminal displays and plays the commentary received. The terminal displays the text commentary received from the server on the screen and plays the audio commentary, allowing the user to understand the real-time situation of a specific target player.

[0768] Example 1

[0769] 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."

[0770] Conventional sports viewing systems have difficulty obtaining real-time information about specific athletes that users are interested in. Furthermore, conventional systems lack sufficient video analysis and commentary generation for specific athletes selected by users, resulting in an unsatisfactory sports viewing experience for users. Furthermore, the mechanisms for providing generated information to user devices in real time are also inadequate.

[0771] 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.

[0772] In this invention, the server includes means for receiving and analyzing game footage in real time, means for identifying specific target players based on the analysis results, means for extracting information about the identified specific target players, means for generating a commentary in natural language based on the extracted information, means for converting the generated natural language commentary into audio data, and means for transmitting the generated commentary to a user terminal, thereby enabling users to obtain information about specific players in real time and significantly improving their viewing experience.

[0773] "Game video" refers to video data that records the state of a sports game.

[0774] "Means for receiving and analyzing in real time" refers to a combination of hardware and software for receiving game footage in real time via a network and sequentially analyzing the footage.

[0775] The "means for identifying a specific target player based on the analysis results" refers to an algorithm and program for identifying a specific player from the analyzed game video data.

[0776] The "means for extracting information about the identified specific target player" is a program for extracting related information such as the position information and movements of the specific player, and their positional relationship with other players, from the analysis results.

[0777] The "means for generating commentary in natural language" refers to an algorithm and program for generating commentary in text format based on the extracted information.

[0778] The "means for converting the commentary in natural language into voice data" is a voice synthesis technology for converting the generated commentary in text format into voice data.

[0779] "Means for transmitting to user terminal" refers to network technologies and protocols for transmitting the generated commentary to the user's device in real time.

[0780] The "means for the user to select a specific target player" refers to an interface for the user to select a favorite player and a program for confirming the selection.

[0781] The "means for transmitting selected information to the server" refers to a network technology for transmitting information about the specific target player selected by the user to the server.

[0782] The "means for generating a request including identification information of a selected specific target player" is a program for generating a data request including identification information about the selected player.

[0783] The "means for transmitting the generated request to the server" refers to a network technology for transmitting the generated request to the server.

[0784] The "means for receiving and displaying the live commentary from the server" refers to a program and interface for receiving the live commentary sent from the server at the user terminal and displaying it on the screen.

[0785] "Means for generating information in natural language using a generative AI model" means a program that uses a generative AI model (e.g., GPT, BERT) to generate commentary in natural language based on extracted information.

[0786] The present invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system combines a user terminal, a server, and a generative AI model to analyze game footage in real time and provide information.

[0787] Users access the system through a smartphone application. When the application is launched, the user can access an interface for selecting a favorite player. When the user selects a specific player (e.g., "Player A") from the player list and confirms the selection, the user's device acquires this information. The device generates a request including the identification information (player ID) of the selected specific target player and sends it to the server. This request is sent using the HTTP POST method.

[0788] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API) and uses a GPU to perform high-speed image analysis. The software used includes generative AI models (e.g., YOLO and OpenPose). Using these models, the server analyzes the game footage frame by frame to detect the positions and movements of specific target players. For example, in a soccer game, the analysis identifies the position and actions (e.g., dribbling, shooting, passing) of "Player A."

[0789] The server extracts information related to "Player A" from the analysis results. During this extraction process, "Player A's" location information, actions, and surrounding circumstances are recorded in detail. This information is used to grasp the data the user wants to know in real time.

[0790] The server generates a commentary in natural language based on the extracted information. A generative AI model (e.g., GPT-3 or BERT) is used for natural language generation. By inputting a prompt sentence such as "Player A is dribbling," the generative AI model generates specific commentary text such as "Player A is dribbling forward, dodging the defender!" This text commentary is then converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly).

[0791] The generated commentary (text and audio data) is sent from the server to the user's device. WebSocket and HTTP / 2 protocols are used to transmit data in real time. The user's device immediately processes the received commentary data. The received commentary text is displayed on the display, and the audio commentary is played through the speaker.

[0792] As a concrete example, consider the case where a user selects a soccer player, "Player A," as their favorite player. The user selects "Player A" using a smartphone app, and that information is sent to the server. The server receives the game footage, analyzes it using a generative AI model, and identifies the scene where "Player A" is dribbling in front of the goal. As a result, a commentary such as "Player A is approaching the goal!" is generated and immediately sent to the user's smartphone. The user's smartphone displays this commentary and simultaneously plays it back as audio.

[0793] Examples of prompts:

[0794] "Player A starts dribbling at 15:42"

[0795] "Player A is dribbling forward past the defender!"

[0796] As described above, the system of the present invention allows users to obtain real-time information specific to a particular athlete, greatly enhancing the sports viewing experience.

[0797] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0798] Step 1:

[0799] User-defined favorite players

[0800] The user launches the smartphone application and sets their favorite player. Specifically, they select a specific target player (for example, "Player A") from the player list in the application and confirm the selection. The input data is the player list and player ID, and the output data is the identification information (player ID) of the selected player. The user selects "Player A," and that ID is stored within the application.

[0801] Step 2:

[0802] Sending a request from the device to the server

[0803] The terminal generates a request including the identification information (player ID) of the specific target player selected by the user. This request is sent to the server using the HTTP POST method. The input data is the player ID, and the output data is the request sent to the server. The terminal converts the player ID into a JSON object and sends it to the URL https: / / api.example.com / select-player.

[0804] Step 3:

[0805] Real-time analysis of game footage by server

[0806] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API). It uses a GPU to process the video data frame by frame and analyzes player movements using a generative AI model (e.g., YOLO or OpenPose). The input data is the live stream video, and the output data is player position and movement information. The server analyzes the game footage frame by frame to identify the position and movement of "Player A."

[0807] Step 4:

[0808] Server extracts information about specific target players

[0809] The server extracts information related to a specific target player ("Player A") from the analysis results. The input data is the frame analysis results, and the output data is "Player A's" position information, actions, and surrounding circumstances. The server records this information in a database, allowing users to grasp the information they want to know in real time.

[0810] Step 5:

[0811] Server-based commentary generation

[0812] The server generates commentary in natural language based on the extracted information. Using a generative AI model (e.g., GPT-3 or BERT), a prompt sentence is input and specific commentary text is generated. The input data is the extracted player information, and the output data is commentary text in natural language. For example, the server generates commentary text such as, "Player A is dribbling forward, dodging the defender!"

[0813] Step 6:

[0814] Live audio conversion by the server

[0815] The generated commentary text is converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The input data is the commentary text, and the output data is audio data. The server converts the generated commentary text into an audio file.

[0816] Step 7:

[0817] Live data transmission from the server to the device

[0818] The server transmits the generated commentary text and audio data to the user terminal in real time. The input data is the commentary text and audio data, and the output data is the transmitted commentary data. This transmission is performed using the WebSocket or HTTP / 2 protocol.

[0819] Step 8:

[0820] Live display and audio playback on the device

[0821] The device processes the commentary data received from the server, displays the text on the display, and plays the audio from the speaker. The input data is the commentary text and audio data, and the output data is the displayed text and the played audio. For example, the commentary text "Player A is closing in on the goal!" is displayed on the screen, and the content is played back as audio.

[0822] (Application example 1)

[0823] 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."

[0824] In conventional sports commentary, it was difficult to obtain detailed information about specific players in real time. In addition, users were unable to receive timely information specific to their favorite players, and were only provided with general information that was in line with the progress of the entire game. This limited users' sports viewing experience, preventing them from enjoying the excitement of focusing on the play of their favorite players in real time.

[0825] 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.

[0826] In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about specific target players based on the analysis results, means for generating commentary about the specific target players based on the extracted information, means for transmitting the generated commentary to a user terminal, and means for displaying and playing back the received commentary on the user terminal as audio. This allows users to receive detailed information about the play of their favorite players in real time, providing a more exciting sports viewing experience through both their eyes and their ears.

[0827] "Game footage" is video data captured during a sporting event.

[0828] "Receiving in real time" means obtaining data without delay while the match is in progress.

[0829] "Means for analysis" refers to a device or program that can extract the movements and positional information of specific players from the received game footage.

[0830] A "specific target athlete" is a specific athlete designated by the user.

[0831] "Means for extracting information" refers to devices or programs that extract data related to specific target players from the analysis results of game footage.

[0832] "Means for generating commentary" refers to a device or program that creates commentary in natural language from the extracted player information.

[0833] A "user terminal" is an electronic device used by a user, and examples include smartphones and tablets.

[0834] The "transmitting means" refers to a communication device or program for delivering the generated commentary to the user terminal.

[0835] "Means for displaying and playing audio" refers to a device or program that displays the received commentary on a screen and plays the audio through speakers or headphones.

[0836] "User-selected means" refers to the interface or operation method that allows the user to specify their favorite player.

[0837] "Means for sending to server" refers to a communication device or program for sending the information selected by the user to the central server.

[0838] "Natural language commentary" is commentary that expresses information about a specific player in natural language that is easy for people to understand.

[0839] "Means for converting into audio data" refers to a device or program that converts the generated natural language text into an audio file.

[0840] This invention is a system that provides real-time commentary specific to a specific athlete. This system provides users with detailed information such as the movements and positions of the specific athlete in real time, thereby improving the sports viewing experience. Specific embodiments of this system are described below.

[0841] System Configuration

[0842] The server includes the following means:

[0843] 1. Means of receiving and analyzing game footage in real time:

[0844] The server receives live-streamed game footage in real time and uses deep learning models such as TensorFlow for video analysis.

[0845] 2. Methods for extracting information on specific target players based on the analysis results:

[0846] The received video is analyzed frame by frame to extract the movement and position information of specific target players, which includes player identification and action recognition.

[0847] 3. A means for generating commentary about a specific target player based on the extracted information:

[0848] Based on the extracted information, a live commentary text is generated using natural language processing technology, using generative AI models such as Transformers for natural language generation.

[0849] 4. Means for transmitting the generated commentary to the user terminal:

[0850] The generated text and voice data are sent to the user's device using real-time data transfer technology (e.g., WebSocket) over the Internet.

[0851] The user terminal includes the following means:

[0852] 1. Means for displaying and playing back received commentary:

[0853] The user device (e.g., a smartphone) displays the received commentary text on its screen and plays the audio data through speakers or headphones using a voice playback and speech synthesis library such as pygame or gTTS.

[0854] 2. How users select specific target players:

[0855] The user selects their favorite player through the application, and the selection information is sent to the server, where the player is set as a specific target player.

[0856] Specific examples

[0857] Specific examples are given below.

[0858] Let's take the example of a scenario where a user is using "Favorite Player Live Assist" on their smartphone to watch a soccer match. The user selects Player A from the application's list and sets him as the specific target player. Meanwhile, the server analyzes the game footage in real time and identifies Player A's movements. For example, when it captures a scene where Player A has the ball and begins to dribble, it generates a commentary such as, "Player A is dribbling forward, dodging the defender." This commentary is immediately sent to the smartphone and displayed on the screen, while also being played back as audio using speech synthesis technology.

[0859] Example prompts for generative AI models

[0860] "Player A is dribbling forward past the defender. Please provide a commentary of the game situation."

[0861] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0862] Step 1:

[0863] The user selects a specific target player

[0864] The user launches the application and selects a specific target player from a list. The user selects their favorite player from the player list (loaded from the database) by tapping on it. As input, the player's identification information is generated and sent to the server. As output, the player ID is selected and sent to the server.

[0865] Step 2:

[0866] Send the selected information to the server

[0867] The terminal sends the identification information of the specific target player selected by the user to the server. The server receives the information and sets the specified player as the specific target player. The player ID is sent from the terminal as input, and the player information is saved in the database in the server as output.

[0868] Step 3:

[0869] The server receives and analyzes the match footage in real time.

[0870] The server receives live-streamed game footage in real time and analyzes it using a deep learning model (such as TensorFlow). Live video data is given as input, and frame-by-frame analysis results are output. Specifically, player movements and position information are identified for each frame.

[0871] Step 4:

[0872] Extract information on specific target players based on the analysis results

[0873] The server extracts information related to a specific target player from the analysis results. The analyzed data of all players is given as input, and data showing the actions and positions of the specific target player is obtained as output. Specifically, the player's actions (e.g., dribbling, passing, shooting) and current position information are extracted.

[0874] Step 5:

[0875] Generate commentary based on the extracted information

[0876] Based on the extracted information, the server uses a generative AI model (e.g., GPT-3) to generate commentary text in natural language. The input is given as movement data of a specific target player, and the output is generated commentary text. Specifically, the generative AI creates text using the prompt sentence, "Player A is dribbling forward, dodging the defender. Please provide commentary on the situation in the game."

[0877] Step 6:

[0878] Send the generated commentary to the user's device

[0879] The server transmits the generated commentary text and audio data to the user terminal. The commentary text and audio files are stored in the server as input, and are transmitted to the user terminal as output. Specifically, the data is transmitted in real time via the Internet.

[0880] Step 7:

[0881] The user device displays the commentary and plays it back with audio.

[0882] The user device displays the received commentary text on the screen and plays the audio data through speakers or headphones. The input is the received text and audio file, and the output is the display of text on the screen and audio playback. The user can follow the movements of a specific player in real time through both sight and sound.

[0883] 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.

[0884] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete), and further combines it with an emotion engine that recognizes the user's emotions. How the program of this system works will be explained using concrete examples.

[0885] Setting favorite players on user devices

[0886] To use the system, a user must first select a specific target player. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the smartphone app and selects "Player A" from the player list.

[0887] The user terminal acquires the selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[0888] Real-time analysis of game footage by server

[0889] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[0890] Identifying information about a specific target player by the server

[0891] The server extracts information related to the specific target player from the analysis results. During this process, the key actions of the specific target player (e.g., dribbling, shooting, passing), their position, and the surrounding situation are precisely recorded. This allows users to obtain the information they want in real time.

[0892] Server-based commentary generation

[0893] Based on the extracted information, the server generates commentary in natural language. This commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[0894] Transmission of commentary generated by the server and display on the user's device

[0895] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to keep track of the latest status of a specific target player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the same commentary in audio.

[0896] Recognizing user emotions with an emotion engine

[0897] The user device is equipped with an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the user's facial expressions and voice tone in real time to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[0898] Adjusting commentary based on emotion information

[0899] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to make it even more exciting. A rich commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!"

[0900] This system allows users to understand the status of specific athletes in real time and receive commentary that is customized to their individual emotions, eliminating frustrations with traditional sports commentary and significantly improving the user's sports viewing experience.

[0901] The processing flow will be explained below.

[0902] Step 1:

[0903] The user launches the application and selects a specific player. The user selects "Player A" from the player list in the app.

[0904] Step 2:

[0905] The terminal transmits the selection information to the server, and the terminal generates a request including the identification information of the specific target player and transmits it to the server.

[0906] Step 3:

[0907] The server receives live-streamed game footage in real time and stores the game footage data stream in a buffer.

[0908] Step 4:

[0909] The server analyzes the game footage frame by frame, and the server's generative AI model processes each frame to identify the players' positions and movements.

[0910] Step 5:

[0911] The server extracts information about the specific target player, and extracts information related to the specific target player from the analysis results (e.g., position data, ball possession status, actions).

[0912] Step 6:

[0913] The server generates commentary based on the extracted information, using natural language generation technology to generate commentary such as, "Player A is dribbling forward, dodging the defender!"

[0914] Step 7:

[0915] The server converts the generated commentary into audio data, and the generated text commentary is converted into audio data using voice synthesis technology.

[0916] Step 8:

[0917] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[0918] Step 9:

[0919] The device displays and plays the commentary it receives. The device displays the received text commentary on the screen and plays the received audio commentary.

[0920] Step 10:

[0921] The device's emotion engine recognizes the user's emotions by analyzing the user's facial expressions and tone of voice to identify the user's emotional state.

[0922] Step 11:

[0923] The device transmits emotion information to the server. The emotion engine transmits the user's emotion information identified by the device to the server.

[0924] Step 12:

[0925] The server adjusts the commentary based on the emotional information. The server receives the user's emotional state and dynamically adjusts the commentary content based on the information.

[0926] Step 13:

[0927] The server transmits the adjusted commentary to the user terminal, and the server transmits the adjusted commentary data to the user terminal again.

[0928] Step 14:

[0929] The terminal displays and plays the adjusted commentary. The terminal displays the received adjusted commentary on the screen and plays it back as audio.

[0930] Example 2

[0931] 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."

[0932] Conventional sports commentary systems make it difficult for users to receive real-time information about specific athletes, and they are unable to provide customized commentary that takes into account the user's emotional state. This results in a standardized viewing experience for users, and makes it difficult to meet individual needs.

[0933] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for recognizing the user's emotional state, means for dynamically adjusting the commentary content based on the recognized emotional information, and means for transmitting the generated commentary to the user terminal. This not only enables the user to receive detailed real-time information about a specific player, but also makes it possible to provide a commentary customized according to the user's emotions.

[0934] "Game footage" refers to video data that records the progress of a sports game in real time.

[0935] The "analysis means" refers to a technology or device that analyzes the input game video and extracts important information from each frame.

[0936] "Specific target athlete" refers to a specific athlete selected by the user.

[0937] "Means for extracting information" refers to a technique or method for extracting data (position information, movements, etc.) related to a specific target player from the analysis results.

[0938] The "means for generating commentary" is a technique or method for describing the actions of a specific target player and the situation of the game in natural language based on the extracted data.

[0939] "User's emotional state" refers to the emotional state obtained by analyzing the user's facial expression, voice tone, and other biometric information.

[0940] The "means for recognizing emotional states" refers to a technology or device that analyzes the user's facial expressions and tone of voice to identify emotions.

[0941] A "means for dynamically adjusting commentary content" is a technique or method for changing commentary content in real time based on recognized emotion information.

[0942] "User terminal" refers to a computer, smartphone, tablet, or other mobile device used by a User.

[0943] The "generated commentary" is text and audio data of a narration generated based on the progress of the match.

[0944] The "transmitting means" refers to a technique or method for transmitting the generated commentary to a user terminal.

[0945] This invention is a system that allows users to receive real-time commentary on specific athletes (hereinafter referred to as "specific target athletes") and further customizes the commentary content by recognizing the user's emotions. The program processing of this system is explained in detail below.

[0946] To use the system, a user must first select a specific target player. The user launches the smartphone app and selects a specific target player from the player list. For example, the user selects "Player A." At this time, the smartphone generates a request including the player's identification information and sends it to the server.

[0947] The server receives this request and uses a generative AI model such as YOLOv5 to analyze each frame of the live-streamed game footage in real time, identifying the position and movement of specific target players and extracting important actions (e.g., dribbling, shooting, passing).

[0948] Based on the extracted information, the server generates commentary for the specific target player using a natural language generation model such as GPT-3. For example, a commentary such as "Player A is dribbling forward, dodging the defender!" is generated. This commentary is then converted into audio data using voice synthesis technology such as Google Cloud Text-to-Speech.

[0949] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays this commentary in real time and plays the audio. For example, the user's smartphone might display the text "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[0950] Furthermore, the user device is equipped with an emotion engine that analyzes the user's facial expressions and voice tone in real time to identify their emotional state. The emotion engine uses TensorFlow and other tools to recognize the user's surprise, excitement, joy, etc., and sends this information to the server. The server receives the user's emotional information and, for example, if the user is very excited, adjusts the commentary to make it more exciting. Commentary that corresponds to the user's emotional state is generated, such as "Player A is showing off some amazing dribbling!"

[0951] This system not only allows users to keep up with the latest game progress of specific players in real time, but also allows them to receive commentary that is customized to their emotions at the time. This technology meets individual needs that traditional sports commentary could not meet, significantly improving the user's viewing experience.

[0952] Examples of specific examples and prompts

[0953] For example, if a user "Tanaka" has set a specific target player "Player A" on his smartphone and is watching a soccer game, and Tanaka looks surprised when "Player A" shoots the ball:

[0954] 1. The user device detects Tanaka's surprised expression, and the emotion engine analyzes it.

[0955] 2. The emotion engine sends the recognized emotion information to the server.

[0956] 3. The server adjusts the commentary based on this information, generating a commentary such as "Player A took a great shot!"

[0957] 4. This live commentary will be displayed in real time on Tanaka's smartphone and played back as audio.

[0958] Example prompt sentence:

[0959] "Generate a commentary of a game played by Player A. If the user seems excited, make the commentary even more exciting."

[0960] The above is a specific embodiment of the present invention, which allows users to enjoy detailed game situations of specific target players and commentary customized to their own emotions in real time.

[0961] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0962] Program processing flow

[0963] Step 1: User's Favourite Player

[0964] 1. The user launches the smartphone app and selects a specific target player from the player list.

[0965] Input: User selects a player within the app

[0966] Output: Identification information of the specific target player is generated.

[0967] Specific operation: When the user selects "Player A" on the app screen, the device generates the identification information of the specific target player (e.g., { "player_id": "A123"}).

[0968] 2. The terminal generates a request including the selection information and sends it to the server.

[0969] Input: Identification information of specific target player

[0970] Output: The request sent to the server

[0971] Specific operation: The device sends a request containing the device-generated identification information to the server in the following format: { "action": "select_player", "player_id": "A123"}

[0972] Step 2: Real-time analysis of game footage by the server

[0973] 1. The server receives live footage of the match.

[0974] Input: Live video stream

[0975] Output: Image data for each frame

[0976] Specific operation: The server uses tools such as FFmpeg to divide the live video into frames and generate image data for each frame.

[0977] 2. The server analyzes each frame using a generative AI model to detect the position and movement of specific target players.

[0978] Input: Frame-by-frame image data

[0979] Output: Position and action information of specific target players

[0980] Specific behavior: Using a generative AI model such as YOLOv5, each frame is analyzed to detect the position (e.g., coordinate information) and actions (e.g., dribbling, shooting) of specific target players.

[0981] Step 3: The server identifies information about the specific target player.

[0982] 1. The server extracts important actions of specific target players from the analysis results.

[0983] Input: Detected location and action information

[0984] Output: List of important actions of a specific target player

[0985] Specific actions: Extract the main actions of a specific target player from the analysis results (e.g., { "time": "10:12", "action": "dribble"}) and save them in list format.

[0986] Step 4: Server-generated commentary

[0987] 1. The server generates commentary text based on the extracted information.

[0988] Input: List of important actions for a specific target player

[0989] Output: Commentary text

[0990] Specific Actions: Use generative AI models such as GPT-3 to generate commentary text based on specific player actions and game situations (e.g., "Player A is dribbling forward, dodging the defender!").

[0991] 2. The server converts the generated text into audio data.

[0992] Input: Play-by-play text

[0993] Output: Audio data

[0994] Specific operation: The generated commentary text is converted into audio data using Google Cloud Text-to-Speech or similar.

[0995] Step 5: Transmission of the generated commentary by the server and display on the user's device

[0996] 1. The server sends the generated commentary (text and audio data) to the user terminal.

[0997] Input: Live text and audio data

[0998] Output: Live data sent to the user's device

[0999] Specific operation: A network communication protocol (e.g., HTTP, WebSocket) is used to send live data to the user terminal.

[1000] 2. The device displays the received commentary in real time and plays it back as audio.

[1001] Input: Live data from the server

[1002] Output: Running text displayed on the screen and audio played back

[1003] Specific operation: The received commentary text is displayed on the device screen and the audio data is played back.

[1004] Step 6: Recognizing user emotions with the emotion engine

[1005] 1. The device collects the user's facial expressions and voice tone in real time.

[1006] Input: User's facial and voice data

[1007] Output: Collected real-time emotion data

[1008] Specific operation: The device's camera and microphone are used to collect the user's facial expressions and voice tone in real time.

[1009] 2. The device analyzes the collected emotional data and identifies the user's emotional state.

[1010] Input: Collected emotion data

[1011] Output: Parsed emotional state information

[1012] Specific operation: Analyze the user's emotional state (e.g., surprise, excitement) using an emotion recognition engine such as TensorFlow.

[1013] 3. The device sends the analyzed emotion information to the server.

[1014] Input: Parsed emotional state information

[1015] Output: Emotion information sent to the server

[1016] Specific operation: A network communication protocol is used to transmit emotional state information to the server.

[1017] Step 7: Adjusting commentary based on emotional information

[1018] 1. The server receives the user's emotional information and adjusts the commentary content based on that emotional information.

[1019] Input: User's emotional information

[1020] Output: Adjusted commentary text

[1021] Specific operation: Based on the user's emotional information, the commentary text is regenerated using a generative AI model such as GPT-3 and the content is adjusted.

[1022] 2. The server sends the adjusted commentary back to the user's device.

[1023] Input: Adjusted play-by-play text

[1024] Output: Adjusted live data sent to user device

[1025] Specific operation: The adjusted live data is sent back to the user's terminal, and the terminal displays and plays it.

[1026] The above are the specific processing steps of the system program, and this series of processes enables users to receive real-time information on specific players and customized commentary according to their emotions.

[1027] (Application example 2)

[1028] 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."

[1029] Conventional sports commentary systems lacked commentary tailored to specific athletes, making it difficult to provide information tailored to users' interests. Furthermore, they lacked commentary tailored to the user's emotions, resulting in a limited viewing experience and low satisfaction. Furthermore, they lacked the ability to recognize emotions in real time and dynamically adjust commentary based on those emotions, making it difficult to provide an optimal viewing experience for each user.

[1030] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game video in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for transmitting the generated commentary to a user terminal, means for recognizing and analyzing the user's emotional state, and means for dynamically adjusting the commentary content based on the user's emotional state. This allows the user to receive detailed real-time commentary tailored to a specific player, and further provides a more satisfying viewing experience through commentary customized to the user's emotions.

[1031] 1. "Game Footage" means real-time video data including the plays and player actions during a game.

[1032] 2. "Analyzing" means automatically analyzing received game video data using a computer model to extract specific information.

[1033] 3. "Specific target player" refers to an individual player who is of interest to the user and for whom the system will specialize in analyzing and providing information.

[1034] 4. "Extracting information" refers to extracting important movement and positional information related to a specific player from the real-time analyzed data.

[1035] 5. "Generating commentary" means automatically generating a commentary in text or audio format based on the extracted information.

[1036] 6. "User Terminal" means an electronic device capable of receiving and displaying data and information from the System, including a smartphone or computer.

[1037] 7. "Emotional state" refers to the user's psychological and emotional state as perceived through facial expressions, tone of voice, etc.

[1038] 8. "Recognize and analyze" means detecting the user's emotional state in real time and analyzing that information.

[1039] 9. "Dynamic adjustment of commentary content" means automatically changing the content or tone of commentary in response to the user's perceived emotional state.

[1040] This invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system receives game footage in real time, analyzes it using a generative AI model, extracts information about specific athletes, and generates commentary from that information. The generated commentary is then sent to the user's device and dynamically adjusted according to the user's emotional state.

[1041] The server receives game footage in real time and analyzes it frame by frame. It uses a generative AI model to detect the movements and location information of specific target players. For example, in a soccer game, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[1042] From the analysis results, the server extracts information about the specific target player. In this process, the specific target player's important actions such as dribbling, shooting, passing, etc., as well as their position information, are recorded.

[1043] The server generates a commentary in natural language based on the extracted information. For example, it might say, "Player A is dribbling forward, dodging the defender!" This commentary is also converted into audio data using speech synthesis technology.

[1044] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays it back as audio. This allows the user to keep track of the latest status of a specific target player in real time during a match. For example, a smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the commentary back as audio.

[1045] The user device is equipped with an emotion engine that recognizes the user's emotional state in real time. The emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[1046] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to be even more exciting. Commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!" This allows users to receive a more satisfying and individually customized commentary.

[1047] For example, the following prompt might be generated:

[1048] "Player A is dribbling forward, dodging the defender!" "Player A is closing in on the goal!"

[1049] This system allows users to receive detailed action of specific players in real time and commentary based on their own emotions, dramatically improving the sports viewing experience.

[1050] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1051] Step 1:

[1052] The user starts the application and selects a specific target player. The user then selects their favorite player from the displayed player list and sends that selection to the server. The input is the player list and the user's selection, and the output is the selected player information being sent to the server.

[1053] Step 2:

[1054] The server receives live-streamed game footage in real time. The input is the game video data, and the output is each frame of that video data. The server splits the game video into frames and prepares them for analysis.

[1055] Step 3:

[1056] The server analyzes each frame of the received game video using a generative AI model to detect the movements of specific target players. The input is video data for each frame, and the output is the position information and action data of the specific target player. The server uses the generative AI model to identify important actions such as dribbling, shooting, and passing.

[1057] Step 4:

[1058] The server extracts information related to the specific target player from the analysis results. The input is the analyzed frame data, and the output is the important actions and position information of the specific target player. The server organizes this information as structured data.

[1059] Step 5:

[1060] The server generates natural language commentary based on the extracted information. The input is the key actions and positional information of a specific target player, and the output is commentary text. The server uses a natural language generation model to create commentary such as, "Player A is dribbling forward, dodging the defender!"

[1061] Step 6:

[1062] The server converts the generated commentary text into audio data using speech synthesis technology. The input is the commentary text and the output is audio data. The server then generates the audio file and prepares it to be sent to the user.

[1063] Step 7:

[1064] The server sends the generated commentary (text and audio data) to the user terminal. The input is the commentary text and audio data, and the output is the commentary data received by the user terminal. The user terminal displays the received commentary in real time and starts playing the audio.

[1065] Step 8:

[1066] The user terminal uses an emotion engine to recognize the user's emotional state in real time. The input is the user's facial expression data and voice tone, and the output is the recognized emotional state data. The emotion engine analyzes the user's facial expression and voice tone to identify the emotional state.

[1067] Step 9:

[1068] The server receives the user's emotional state and dynamically adjusts the commentary content based on that information. The input is emotional state data, and the output is adjusted commentary text and audio data. For example, if the user is excited, the commentary content will be changed to something like, "Player A is showing off some amazing dribbling!"

[1069] Step 10:

[1070] The server then sends the adjusted commentary back to the user's device, where the user receives the customized commentary in real time. The input is the adjusted commentary data, and the output is the adjusted commentary information displayed on the user's device, allowing the user to enjoy a highly personalized viewing experience.

[1071] 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.

[1072] 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.

[1073] 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.

[1074] [Fourth embodiment]

[1075] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1076] 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.

[1077] 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).

[1078] 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.

[1079] 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.

[1080] 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).

[1081] 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.

[1082] 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.

[1083] 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.

[1084] 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.

[1085] 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.

[1086] 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.

[1087] 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."

[1088] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete). Below, we will explain how the program of this system works, using concrete examples.

[1089] Setting favorite players on user devices

[1090] To use the system, a user first selects a specific target player to support. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the application on their smartphone and selects "Player A" from the player list.

[1091] The user terminal acquires this selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[1092] Real-time analysis of game footage by server

[1093] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, it detects the movements of each player in each frame and identifies the position and actions of specific target players.

[1094] Identifying information about a specific target player by the server

[1095] The server extracts information related to the specific target player from the analysis results. In this process, the server precisely records the specific target player's important actions (e.g., dribbling, shooting, passing), their position information, and the surrounding situation. This allows users to obtain the information they want in real time.

[1096] Server-based commentary generation

[1097] Based on the extracted information, the server generates a commentary in natural language. The generated commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[1098] Transmission of commentary generated by the server and display on the user's device

[1099] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to check the latest status of their favorite player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[1100] This system allows users to obtain timely information specific to their favorite athletes, eliminating the frustrations of conventional live sports broadcasts. In this way, the present invention significantly improves the user's sports viewing experience.

[1101] The processing flow will be explained below.

[1102] Step 1:

[1103] The user launches the application and selects a specific target player. The user selects their favorite target player from the player list in the application and enters the selection information into the device.

[1104] Step 2:

[1105] The terminal sends the selection information to the server. The terminal generates a request including the identification information (e.g., player ID, name) of the specific target player selected by the user and sends it to the server.

[1106] Step 3:

[1107] The server receives live-streamed game footage in real time. The server continuously receives the game footage data stream and stores it in a buffer.

[1108] Step 4:

[1109] The server analyzes the game footage frame by frame, and the server's generative AI model processes the received footage frame by frame to identify the positions and movements of players in each frame.

[1110] Step 5:

[1111] The server extracts information about the specific target player. The server extracts information related to the specific target player selected by the user (e.g., position data, ball possession status, actions) from the analyzed video data.

[1112] Step 6:

[1113] The server generates a commentary based on the extracted information. The server generates a commentary in natural language using the information on the specific target player, and then converts the commentary into audio data using speech synthesis technology.

[1114] Step 7:

[1115] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[1116] Step 8:

[1117] The terminal displays and plays the commentary received. The terminal displays the text commentary received from the server on the screen and plays the audio commentary, allowing the user to understand the real-time situation of a specific target player.

[1118] Example 1

[1119] 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."

[1120] Conventional sports viewing systems have difficulty obtaining real-time information about specific athletes that users are interested in. Furthermore, conventional systems lack sufficient video analysis and commentary generation for specific athletes selected by users, resulting in an unsatisfactory sports viewing experience for users. Furthermore, the mechanisms for providing generated information to user devices in real time are also inadequate.

[1121] 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.

[1122] In this invention, the server includes means for receiving and analyzing game footage in real time, means for identifying specific target players based on the analysis results, means for extracting information about the identified specific target players, means for generating a commentary in natural language based on the extracted information, means for converting the generated natural language commentary into audio data, and means for transmitting the generated commentary to a user terminal, thereby enabling users to obtain information about specific players in real time and significantly improving their viewing experience.

[1123] "Game video" refers to video data that records the state of a sports game.

[1124] "Means for receiving and analyzing in real time" refers to a combination of hardware and software for receiving game footage in real time via a network and sequentially analyzing the footage.

[1125] The "means for identifying a specific target player based on the analysis results" refers to an algorithm and program for identifying a specific player from the analyzed game video data.

[1126] The "means for extracting information about the identified specific target player" is a program for extracting related information such as the position information and movements of the specific player, and their positional relationship with other players, from the analysis results.

[1127] The "means for generating commentary in natural language" refers to an algorithm and program for generating commentary in text format based on the extracted information.

[1128] The "means for converting the commentary in natural language into voice data" is a voice synthesis technology for converting the generated commentary in text format into voice data.

[1129] "Means for transmitting to user terminal" refers to network technologies and protocols for transmitting the generated commentary to the user's device in real time.

[1130] The "means for the user to select a specific target player" refers to an interface for the user to select a favorite player and a program for confirming the selection.

[1131] The "means for transmitting selected information to the server" refers to a network technology for transmitting information about the specific target player selected by the user to the server.

[1132] The "means for generating a request including identification information of a selected specific target player" is a program for generating a data request including identification information about the selected player.

[1133] The "means for transmitting the generated request to the server" refers to a network technology for transmitting the generated request to the server.

[1134] The "means for receiving and displaying the live commentary from the server" refers to a program and interface for receiving the live commentary sent from the server at the user terminal and displaying it on the screen.

[1135] "Means for generating information in natural language using a generative AI model" means a program that uses a generative AI model (e.g., GPT, BERT) to generate commentary in natural language based on extracted information.

[1136] The present invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system combines a user terminal, a server, and a generative AI model to analyze game footage in real time and provide information.

[1137] Users access the system through a smartphone application. When the application is launched, the user can access an interface for selecting a favorite player. When the user selects a specific player (e.g., "Player A") from the player list and confirms the selection, the user's device acquires this information. The device generates a request including the identification information (player ID) of the selected specific target player and sends it to the server. This request is sent using the HTTP POST method.

[1138] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API) and uses a GPU to perform high-speed image analysis. The software used includes generative AI models (e.g., YOLO and OpenPose). Using these models, the server analyzes the game footage frame by frame to detect the positions and movements of specific target players. For example, in a soccer game, the analysis identifies the position and actions (e.g., dribbling, shooting, passing) of "Player A."

[1139] The server extracts information related to "Player A" from the analysis results. During this extraction process, "Player A's" location information, actions, and surrounding circumstances are recorded in detail. This information is used to grasp the data the user wants to know in real time.

[1140] The server generates a commentary in natural language based on the extracted information. A generative AI model (e.g., GPT-3 or BERT) is used for natural language generation. By inputting a prompt sentence such as "Player A is dribbling," the generative AI model generates specific commentary text such as "Player A is dribbling forward, dodging the defender!" This text commentary is then converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly).

[1141] The generated commentary (text and audio data) is sent from the server to the user's device. WebSocket and HTTP / 2 protocols are used to transmit data in real time. The user's device immediately processes the received commentary data. The received commentary text is displayed on the display, and the audio commentary is played through the speaker.

[1142] As a concrete example, consider the case where a user selects a soccer player, "Player A," as their favorite player. The user selects "Player A" using a smartphone app, and that information is sent to the server. The server receives the game footage, analyzes it using a generative AI model, and identifies the scene where "Player A" is dribbling in front of the goal. As a result, a commentary such as "Player A is approaching the goal!" is generated and immediately sent to the user's smartphone. The user's smartphone displays this commentary and simultaneously plays it back as audio.

[1143] Examples of prompts:

[1144] "Player A starts dribbling at 15:42"

[1145] "Player A is dribbling forward past the defender!"

[1146] As described above, the system of the present invention allows users to obtain real-time information specific to a particular athlete, greatly enhancing the sports viewing experience.

[1147] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1148] Step 1:

[1149] User-defined favorite players

[1150] The user launches the smartphone application and sets their favorite player. Specifically, they select a specific target player (for example, "Player A") from the player list in the application and confirm the selection. The input data is the player list and player ID, and the output data is the identification information (player ID) of the selected player. The user selects "Player A," and that ID is stored within the application.

[1151] Step 2:

[1152] Sending a request from the device to the server

[1153] The terminal generates a request including the identification information (player ID) of the specific target player selected by the user. This request is sent to the server using the HTTP POST method. The input data is the player ID, and the output data is the request sent to the server. The terminal converts the player ID into a JSON object and sends it to the URL https: / / api.example.com / select-player.

[1154] Step 3:

[1155] Real-time analysis of game footage by server

[1156] The server receives game footage in real time from an existing streaming service (e.g., YouTube Live API). It uses a GPU to process the video data frame by frame and analyzes player movements using a generative AI model (e.g., YOLO or OpenPose). The input data is the live stream video, and the output data is player position and movement information. The server analyzes the game footage frame by frame to identify the position and movement of "Player A."

[1157] Step 4:

[1158] Server extracts information about specific target players

[1159] The server extracts information related to a specific target player ("Player A") from the analysis results. The input data is the frame analysis results, and the output data is "Player A's" position information, actions, and surrounding circumstances. The server records this information in a database, allowing users to grasp the information they want to know in real time.

[1160] Step 5:

[1161] Server-based commentary generation

[1162] The server generates commentary in natural language based on the extracted information. Using a generative AI model (e.g., GPT-3 or BERT), a prompt sentence is input and specific commentary text is generated. The input data is the extracted player information, and the output data is commentary text in natural language. For example, the server generates commentary text such as, "Player A is dribbling forward, dodging the defender!"

[1163] Step 6:

[1164] Live audio conversion by the server

[1165] The generated commentary text is converted into audio data using speech synthesis technology (e.g., Google Text-to-Speech or Amazon Polly). The input data is the commentary text, and the output data is audio data. The server converts the generated commentary text into an audio file.

[1166] Step 7:

[1167] Live data transmission from the server to the device

[1168] The server transmits the generated commentary text and audio data to the user terminal in real time. The input data is the commentary text and audio data, and the output data is the transmitted commentary data. This transmission is performed using the WebSocket or HTTP / 2 protocol.

[1169] Step 8:

[1170] Live display and audio playback on the device

[1171] The device processes the commentary data received from the server, displays the text on the display, and plays the audio from the speaker. The input data is the commentary text and audio data, and the output data is the displayed text and the played audio. For example, the commentary text "Player A is closing in on the goal!" is displayed on the screen, and the content is played back as audio.

[1172] (Application example 1)

[1173] 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."

[1174] In conventional sports commentary, it was difficult to obtain detailed information about specific players in real time. In addition, users were unable to receive timely information specific to their favorite players, and were only provided with general information that was in line with the progress of the entire game. This limited users' sports viewing experience, preventing them from enjoying the excitement of focusing on the play of their favorite players in real time.

[1175] 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.

[1176] In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about specific target players based on the analysis results, means for generating commentary about the specific target players based on the extracted information, means for transmitting the generated commentary to a user terminal, and means for displaying and playing back the received commentary on the user terminal as audio. This allows users to receive detailed information about the play of their favorite players in real time, providing a more exciting sports viewing experience through both their eyes and their ears.

[1177] "Game footage" is video data captured during a sporting event.

[1178] "Receiving in real time" means obtaining data without delay while the match is in progress.

[1179] "Means for analysis" refers to a device or program that can extract the movements and positional information of specific players from the received game footage.

[1180] A "specific target athlete" is a specific athlete designated by the user.

[1181] "Means for extracting information" refers to devices or programs that extract data related to specific target players from the analysis results of game footage.

[1182] "Means for generating commentary" refers to a device or program that creates commentary in natural language from the extracted player information.

[1183] A "user terminal" is an electronic device used by a user, and examples include smartphones and tablets.

[1184] The "transmitting means" refers to a communication device or program for delivering the generated commentary to the user terminal.

[1185] "Means for displaying and playing audio" refers to a device or program that displays the received commentary on a screen and plays the audio through speakers or headphones.

[1186] "User-selected means" refers to the interface or operation method that allows the user to specify their favorite player.

[1187] "Means for sending to server" refers to a communication device or program for sending the information selected by the user to the central server.

[1188] "Natural language commentary" is commentary that expresses information about a specific player in natural language that is easy for people to understand.

[1189] "Means for converting into audio data" refers to a device or program that converts the generated natural language text into an audio file.

[1190] This invention is a system that provides real-time commentary specific to a specific athlete. This system provides users with detailed information such as the movements and positions of the specific athlete in real time, thereby improving the sports viewing experience. Specific embodiments of this system are described below.

[1191] System Configuration

[1192] The server includes the following means:

[1193] 1. Means of receiving and analyzing game footage in real time:

[1194] The server receives live-streamed game footage in real time and uses deep learning models such as TensorFlow for video analysis.

[1195] 2. Methods for extracting information on specific target players based on the analysis results:

[1196] The received video is analyzed frame by frame to extract the movement and position information of specific target players, which includes player identification and action recognition.

[1197] 3. A means for generating commentary about a specific target player based on the extracted information:

[1198] Based on the extracted information, a live commentary text is generated using natural language processing technology, using generative AI models such as Transformers for natural language generation.

[1199] 4. Means for transmitting the generated commentary to the user terminal:

[1200] The generated text and voice data are sent to the user's device using real-time data transfer technology (e.g., WebSocket) over the Internet.

[1201] The user terminal includes the following means:

[1202] 1. Means for displaying and playing back received commentary:

[1203] The user device (e.g., a smartphone) displays the received commentary text on its screen and plays the audio data through speakers or headphones using a voice playback and speech synthesis library such as pygame or gTTS.

[1204] 2. How users select specific target players:

[1205] The user selects their favorite player through the application, and the selection information is sent to the server, where the player is set as a specific target player.

[1206] Specific examples

[1207] Specific examples are given below.

[1208] Let's take the example of a scenario where a user is using "Favorite Player Live Assist" on their smartphone to watch a soccer match. The user selects Player A from the application's list and sets him as the specific target player. Meanwhile, the server analyzes the game footage in real time and identifies Player A's movements. For example, when it captures a scene where Player A has the ball and begins to dribble, it generates a commentary such as, "Player A is dribbling forward, dodging the defender." This commentary is immediately sent to the smartphone and displayed on the screen, while also being played back as audio using speech synthesis technology.

[1209] Example prompts for generative AI models

[1210] "Player A is dribbling forward past the defender. Please provide a commentary of the game situation."

[1211] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1212] Step 1:

[1213] The user selects a specific target player

[1214] The user launches the application and selects a specific target player from a list. The user selects their favorite player from the player list (loaded from the database) by tapping on it. As input, the player's identification information is generated and sent to the server. As output, the player ID is selected and sent to the server.

[1215] Step 2:

[1216] Send the selected information to the server

[1217] The terminal sends the identification information of the specific target player selected by the user to the server. The server receives the information and sets the specified player as the specific target player. The player ID is sent from the terminal as input, and the player information is saved in the database in the server as output.

[1218] Step 3:

[1219] The server receives and analyzes the match footage in real time.

[1220] The server receives live-streamed game footage in real time and analyzes it using a deep learning model (such as TensorFlow). Live video data is given as input, and frame-by-frame analysis results are output. Specifically, player movements and position information are identified for each frame.

[1221] Step 4:

[1222] Extract information on specific target players based on the analysis results

[1223] The server extracts information related to a specific target player from the analysis results. The analyzed data of all players is given as input, and data showing the actions and positions of the specific target player is obtained as output. Specifically, the player's actions (e.g., dribbling, passing, shooting) and current position information are extracted.

[1224] Step 5:

[1225] Generate commentary based on the extracted information

[1226] Based on the extracted information, the server uses a generative AI model (e.g., GPT-3) to generate commentary text in natural language. The input is given as movement data of a specific target player, and the output is generated commentary text. Specifically, the generative AI creates text using the prompt sentence, "Player A is dribbling forward, dodging the defender. Please provide commentary on the situation in the game."

[1227] Step 6:

[1228] Send the generated commentary to the user's device

[1229] The server transmits the generated commentary text and audio data to the user terminal. The commentary text and audio files are stored in the server as input, and are transmitted to the user terminal as output. Specifically, the data is transmitted in real time via the Internet.

[1230] Step 7:

[1231] The user device displays the commentary and plays it back with audio.

[1232] The user device displays the received commentary text on the screen and plays the audio data through speakers or headphones. The input is the received text and audio file, and the output is the display of text on the screen and audio playback. The user can follow the movements of a specific player in real time through both sight and sound.

[1233] 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.

[1234] This invention is a system that allows users to receive real-time commentary specific to a specific athlete (hereinafter referred to as a specific target athlete), and further combines it with an emotion engine that recognizes the user's emotions. How the program of this system works will be explained using concrete examples.

[1235] Setting favorite players on user devices

[1236] To use the system, a user must first select a specific target player. The user launches the application and accesses the interface for selecting a specific target player. For example, the user launches the smartphone app and selects "Player A" from the player list.

[1237] The user terminal acquires the selection information and generates a request to send to the server, which includes the identification information of the specific target player, allowing the server to identify the specific target player based on the received information.

[1238] Real-time analysis of game footage by server

[1239] The server receives live-streamed game footage in real time and analyzes it frame by frame using a generative AI model. For example, in a soccer match, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[1240] Identifying information about a specific target player by the server

[1241] The server extracts information related to the specific target player from the analysis results. During this process, the key actions of the specific target player (e.g., dribbling, shooting, passing), their position, and the surrounding situation are precisely recorded. This allows users to obtain the information they want in real time.

[1242] Server-based commentary generation

[1243] Based on the extracted information, the server generates commentary in natural language. This commentary explains the actions of the specific target player and the current game situation. For example, if a specific target player is dribbling, the commentary generated will be, "Player A is dribbling forward, dodging the defender!" In addition, this text commentary is converted into audio data using speech synthesis technology.

[1244] Transmission of commentary generated by the server and display on the user's device

[1245] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays the audio commentary. This allows the user to keep track of the latest status of a specific target player in real time during the game. For example, the user's smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the same commentary in audio.

[1246] Recognizing user emotions with an emotion engine

[1247] The user device is equipped with an emotion engine that recognizes the user's emotional state. The emotion engine analyzes the user's facial expressions and voice tone in real time to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[1248] Adjusting commentary based on emotion information

[1249] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to make it even more exciting. A rich commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!"

[1250] This system allows users to understand the status of specific athletes in real time and receive commentary that is customized to their individual emotions, eliminating frustrations with traditional sports commentary and significantly improving the user's sports viewing experience.

[1251] The processing flow will be explained below.

[1252] Step 1:

[1253] The user launches the application and selects a specific player. The user selects "Player A" from the player list in the app.

[1254] Step 2:

[1255] The terminal transmits the selection information to the server, and the terminal generates a request including the identification information of the specific target player and transmits it to the server.

[1256] Step 3:

[1257] The server receives live-streamed game footage in real time and stores the game footage data stream in a buffer.

[1258] Step 4:

[1259] The server analyzes the game footage frame by frame, and the server's generative AI model processes each frame to identify the players' positions and movements.

[1260] Step 5:

[1261] The server extracts information about the specific target player, and extracts information related to the specific target player from the analysis results (e.g., position data, ball possession status, actions).

[1262] Step 6:

[1263] The server generates commentary based on the extracted information, using natural language generation technology to generate commentary such as, "Player A is dribbling forward, dodging the defender!"

[1264] Step 7:

[1265] The server converts the generated commentary into audio data, and the generated text commentary is converted into audio data using voice synthesis technology.

[1266] Step 8:

[1267] The server transmits the generated commentary to the user terminal, and the server transmits the generated text data and audio data to the user terminal in real time.

[1268] Step 9:

[1269] The device displays and plays the commentary it receives. The device displays the received text commentary on the screen and plays the received audio commentary.

[1270] Step 10:

[1271] The device's emotion engine recognizes the user's emotions by analyzing the user's facial expressions and tone of voice to identify the user's emotional state.

[1272] Step 11:

[1273] The device transmits emotion information to the server. The emotion engine transmits the user's emotion information identified by the device to the server.

[1274] Step 12:

[1275] The server adjusts the commentary based on the emotional information. The server receives the user's emotional state and dynamically adjusts the commentary content based on the information.

[1276] Step 13:

[1277] The server transmits the adjusted commentary to the user terminal, and the server transmits the adjusted commentary data to the user terminal again.

[1278] Step 14:

[1279] The terminal displays and plays the adjusted commentary. The terminal displays the received adjusted commentary on the screen and plays it back as audio.

[1280] Example 2

[1281] 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."

[1282] Conventional sports commentary systems make it difficult for users to receive real-time information about specific athletes, and they are unable to provide customized commentary that takes into account the user's emotional state. This results in a standardized viewing experience for users, and makes it difficult to meet individual needs.

[1283] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game footage in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for recognizing the user's emotional state, means for dynamically adjusting the commentary content based on the recognized emotional information, and means for transmitting the generated commentary to the user terminal. This not only enables the user to receive detailed real-time information about a specific player, but also makes it possible to provide a commentary customized according to the user's emotions.

[1284] "Game footage" refers to video data that records the progress of a sports game in real time.

[1285] The "analysis means" refers to a technology or device that analyzes the input game video and extracts important information from each frame.

[1286] "Specific target athlete" refers to a specific athlete selected by the user.

[1287] "Means for extracting information" refers to a technique or method for extracting data (position information, movements, etc.) related to a specific target player from the analysis results.

[1288] The "means for generating commentary" is a technique or method for describing the actions of a specific target player and the situation of the game in natural language based on the extracted data.

[1289] "User's emotional state" refers to the emotional state obtained by analyzing the user's facial expression, voice tone, and other biometric information.

[1290] The "means for recognizing emotional states" refers to a technology or device that analyzes the user's facial expressions and tone of voice to identify emotions.

[1291] A "means for dynamically adjusting commentary content" is a technique or method for changing commentary content in real time based on recognized emotion information.

[1292] "User terminal" refers to a computer, smartphone, tablet, or other mobile device used by a User.

[1293] The "generated commentary" is text and audio data of a narration generated based on the progress of the match.

[1294] The "transmitting means" refers to a technique or method for transmitting the generated commentary to a user terminal.

[1295] This invention is a system that allows users to receive real-time commentary on specific athletes (hereinafter referred to as "specific target athletes") and further customizes the commentary content by recognizing the user's emotions. The program processing of this system is explained in detail below.

[1296] To use the system, a user must first select a specific target player. The user launches the smartphone app and selects a specific target player from the player list. For example, the user selects "Player A." At this time, the smartphone generates a request including the player's identification information and sends it to the server.

[1297] The server receives this request and uses a generative AI model such as YOLOv5 to analyze each frame of the live-streamed game footage in real time, identifying the position and movement of specific target players and extracting important actions (e.g., dribbling, shooting, passing).

[1298] Based on the extracted information, the server generates commentary for the specific target player using a natural language generation model such as GPT-3. For example, a commentary such as "Player A is dribbling forward, dodging the defender!" is generated. This commentary is then converted into audio data using voice synthesis technology such as Google Cloud Text-to-Speech.

[1299] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays this commentary in real time and plays the audio. For example, the user's smartphone might display the text "Player A is closing in on the goal!" on the screen and also play the commentary as audio.

[1300] Furthermore, the user device is equipped with an emotion engine that analyzes the user's facial expressions and voice tone in real time to identify their emotional state. The emotion engine uses TensorFlow and other tools to recognize the user's surprise, excitement, joy, etc., and sends this information to the server. The server receives the user's emotional information and, for example, if the user is very excited, adjusts the commentary to make it more exciting. Commentary that corresponds to the user's emotional state is generated, such as "Player A is showing off some amazing dribbling!"

[1301] This system not only allows users to keep up with the latest game progress of specific players in real time, but also allows them to receive commentary that is customized to their emotions at the time. This technology meets individual needs that traditional sports commentary could not meet, significantly improving the user's viewing experience.

[1302] Examples of specific examples and prompts

[1303] For example, if a user "Tanaka" has set a specific target player "Player A" on his smartphone and is watching a soccer game, and Tanaka looks surprised when "Player A" shoots the ball:

[1304] 1. The user device detects Tanaka's surprised expression, and the emotion engine analyzes it.

[1305] 2. The emotion engine sends the recognized emotion information to the server.

[1306] 3. The server adjusts the commentary based on this information, generating a commentary such as "Player A took a great shot!"

[1307] 4. This live commentary will be displayed in real time on Tanaka's smartphone and played back as audio.

[1308] Example prompt sentence:

[1309] "Generate a commentary of a game played by Player A. If the user seems excited, make the commentary even more exciting."

[1310] The above is a specific embodiment of the present invention, which allows users to enjoy detailed game situations of specific target players and commentary customized to their own emotions in real time.

[1311] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1312] Program processing flow

[1313] Step 1: User's Favourite Player

[1314] 1. The user launches the smartphone app and selects a specific target player from the player list.

[1315] Input: User selects a player within the app

[1316] Output: Identification information of the specific target player is generated.

[1317] Specific operation: When the user selects "Player A" on the app screen, the device generates the identification information of the specific target player (e.g., { "player_id": "A123"}).

[1318] 2. The terminal generates a request including the selection information and sends it to the server.

[1319] Input: Identification information of specific target player

[1320] Output: The request sent to the server

[1321] Specific operation: The device sends a request containing the device-generated identification information to the server in the following format: { "action": "select_player", "player_id": "A123"}

[1322] Step 2: Real-time analysis of game footage by the server

[1323] 1. The server receives live footage of the match.

[1324] Input: Live video stream

[1325] Output: Image data for each frame

[1326] Specific operation: The server uses tools such as FFmpeg to divide the live video into frames and generate image data for each frame.

[1327] 2. The server analyzes each frame using a generative AI model to detect the position and movement of specific target players.

[1328] Input: Frame-by-frame image data

[1329] Output: Position and action information of specific target players

[1330] Specific behavior: Using a generative AI model such as YOLOv5, each frame is analyzed to detect the position (e.g., coordinate information) and actions (e.g., dribbling, shooting) of specific target players.

[1331] Step 3: The server identifies information about the specific target player.

[1332] 1. The server extracts important actions of specific target players from the analysis results.

[1333] Input: Detected location and action information

[1334] Output: List of important actions of a specific target player

[1335] Specific actions: Extract the main actions of a specific target player from the analysis results (e.g., { "time": "10:12", "action": "dribble"}) and save them in list format.

[1336] Step 4: Server-generated commentary

[1337] 1. The server generates commentary text based on the extracted information.

[1338] Input: List of important actions for a specific target player

[1339] Output: Commentary text

[1340] Specific Actions: Use generative AI models such as GPT-3 to generate commentary text based on specific player actions and game situations (e.g., "Player A is dribbling forward, dodging the defender!").

[1341] 2. The server converts the generated text into audio data.

[1342] Input: Play-by-play text

[1343] Output: Audio data

[1344] Specific operation: The generated commentary text is converted into audio data using Google Cloud Text-to-Speech or similar.

[1345] Step 5: Transmission of the generated commentary by the server and display on the user's device

[1346] 1. The server sends the generated commentary (text and audio data) to the user terminal.

[1347] Input: Live text and audio data

[1348] Output: Live data sent to the user's device

[1349] Specific operation: A network communication protocol (e.g., HTTP, WebSocket) is used to send live data to the user terminal.

[1350] 2. The device displays the received commentary in real time and plays it back as audio.

[1351] Input: Live data from the server

[1352] Output: Running text displayed on the screen and audio played back

[1353] Specific operation: The received commentary text is displayed on the device screen and the audio data is played back.

[1354] Step 6: Recognizing user emotions with the emotion engine

[1355] 1. The device collects the user's facial expressions and voice tone in real time.

[1356] Input: User's facial and voice data

[1357] Output: Collected real-time emotion data

[1358] Specific operation: The device's camera and microphone are used to collect the user's facial expressions and voice tone in real time.

[1359] 2. The device analyzes the collected emotional data and identifies the user's emotional state.

[1360] Input: Collected emotion data

[1361] Output: Parsed emotional state information

[1362] Specific operation: Analyze the user's emotional state (e.g., surprise, excitement) using an emotion recognition engine such as TensorFlow.

[1363] 3. The device sends the analyzed emotion information to the server.

[1364] Input: Parsed emotional state information

[1365] Output: Emotion information sent to the server

[1366] Specific operation: A network communication protocol is used to transmit emotional state information to the server.

[1367] Step 7: Adjusting commentary based on emotional information

[1368] 1. The server receives the user's emotional information and adjusts the commentary content based on that emotional information.

[1369] Input: User's emotional information

[1370] Output: Adjusted commentary text

[1371] Specific operation: Based on the user's emotional information, the commentary text is regenerated using a generative AI model such as GPT-3 and the content is adjusted.

[1372] 2. The server sends the adjusted commentary back to the user's device.

[1373] Input: Adjusted play-by-play text

[1374] Output: Adjusted live data sent to user device

[1375] Specific operation: The adjusted live data is sent back to the user's terminal, and the terminal displays and plays it.

[1376] The above are the specific processing steps of the system program, and this series of processes enables users to receive real-time information on specific players and customized commentary according to their emotions.

[1377] (Application example 2)

[1378] 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."

[1379] Conventional sports commentary systems lacked commentary tailored to specific athletes, making it difficult to provide information tailored to users' interests. Furthermore, they lacked commentary tailored to the user's emotions, resulting in a limited viewing experience and low satisfaction. Furthermore, they lacked the ability to recognize emotions in real time and dynamically adjust commentary based on those emotions, making it difficult to provide an optimal viewing experience for each user.

[1380] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and analyzing game video in real time, means for extracting information about the specific target player based on the analysis results, means for generating a commentary about the specific target player based on the extracted information, means for transmitting the generated commentary to a user terminal, means for recognizing and analyzing the user's emotional state, and means for dynamically adjusting the commentary content based on the user's emotional state. This allows the user to receive detailed real-time commentary tailored to a specific player, and further provides a more satisfying viewing experience through commentary customized to the user's emotions.

[1381] 1. "Game Footage" means real-time video data including the plays and player actions during a game.

[1382] 2. "Analyzing" means automatically analyzing received game video data using a computer model to extract specific information.

[1383] 3. "Specific target player" refers to an individual player who is of interest to the user and for whom the system will specialize in analyzing and providing information.

[1384] 4. "Extracting information" refers to extracting important movement and positional information related to a specific player from the real-time analyzed data.

[1385] 5. "Generating commentary" means automatically generating a commentary in text or audio format based on the extracted information.

[1386] 6. "User Terminal" means an electronic device capable of receiving and displaying data and information from the System, including a smartphone or computer.

[1387] 7. "Emotional state" refers to the user's psychological and emotional state as perceived through facial expressions, tone of voice, etc.

[1388] 8. "Recognize and analyze" means detecting the user's emotional state in real time and analyzing that information.

[1389] 9. "Dynamic adjustment of commentary content" means automatically changing the content or tone of commentary in response to the user's perceived emotional state.

[1390] This invention is a system that allows users to receive real-time commentary tailored to specific athletes. The system receives game footage in real time, analyzes it using a generative AI model, extracts information about specific athletes, and generates commentary from that information. The generated commentary is then sent to the user's device and dynamically adjusted according to the user's emotional state.

[1391] The server receives game footage in real time and analyzes it frame by frame. It uses a generative AI model to detect the movements and location information of specific target players. For example, in a soccer game, the server detects the movements of each player in each frame and identifies the position and actions of specific target players.

[1392] From the analysis results, the server extracts information about the specific target player. In this process, the specific target player's important actions such as dribbling, shooting, passing, etc., as well as their position information, are recorded.

[1393] The server generates a commentary in natural language based on the extracted information. For example, it might say, "Player A is dribbling forward, dodging the defender!" This commentary is also converted into audio data using speech synthesis technology.

[1394] The generated commentary (text and audio data) is sent from the server to the user's device. The user's device displays the received commentary in real time and plays it back as audio. This allows the user to keep track of the latest status of a specific target player in real time during a match. For example, a smartphone might display text such as "Player A is closing in on the goal!" on the screen and play the commentary back as audio.

[1395] The user device is equipped with an emotion engine that recognizes the user's emotional state in real time. The emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state. For example, if the user has a surprised expression, the emotion engine sends that information to the server.

[1396] The server receives the user's emotional state and dynamically adjusts the commentary based on that information. For example, if the user is very excited, the commentary will be adjusted to be even more exciting. Commentary that reflects the user's emotions will be generated, such as "Player A is showing off some incredible dribbling!" This allows users to receive a more satisfying and individually customized commentary.

[1397] For example, the following prompt might be generated:

[1398] "Player A is dribbling forward, dodging the defender!" "Player A is closing in on the goal!"

[1399] This system allows users to receive detailed action of specific players in real time and commentary based on their own emotions, dramatically improving the sports viewing experience.

[1400] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1401] Step 1:

[1402] The user starts the application and selects a specific target player. The user then selects their favorite player from the displayed player list and sends that selection to the server. The input is the player list and the user's selection, and the output is the selected player information being sent to the server.

[1403] Step 2:

[1404] The server receives live-streamed game footage in real time. The input is the game video data, and the output is each frame of that video data. The server splits the game video into frames and prepares them for analysis.

[1405] Step 3:

[1406] The server analyzes each frame of the received game video using a generative AI model to detect the movements of specific target players. The input is video data for each frame, and the output is the position information and action data of the specific target player. The server uses the generative AI model to identify important actions such as dribbling, shooting, and passing.

[1407] Step 4:

[1408] The server extracts information related to the specific target player from the analysis results. The input is the analyzed frame data, and the output is the important actions and position information of the specific target player. The server organizes this information as structured data.

[1409] Step 5:

[1410] The server generates natural language commentary based on the extracted information. The input is the key actions and positional information of a specific target player, and the output is commentary text. The server uses a natural language generation model to create commentary such as, "Player A is dribbling forward, dodging the defender!"

[1411] Step 6:

[1412] The server converts the generated commentary text into audio data using speech synthesis technology. The input is the commentary text and the output is audio data. The server then generates the audio file and prepares it to be sent to the user.

[1413] Step 7:

[1414] The server sends the generated commentary (text and audio data) to the user terminal. The input is the commentary text and audio data, and the output is the commentary data received by the user terminal. The user terminal displays the received commentary in real time and starts playing the audio.

[1415] Step 8:

[1416] The user terminal uses an emotion engine to recognize the user's emotional state in real time. The input is the user's facial expression data and voice tone, and the output is the recognized emotional state data. The emotion engine analyzes the user's facial expression and voice tone to identify the emotional state.

[1417] Step 9:

[1418] The server receives the user's emotional state and dynamically adjusts the commentary content based on that information. The input is emotional state data, and the output is adjusted commentary text and audio data. For example, if the user is excited, the commentary content will be changed to something like, "Player A is showing off some amazing dribbling!"

[1419] Step 10:

[1420] The server then sends the adjusted commentary back to the user's device, where the user receives the customized commentary in real time. The input is the adjusted commentary data, and the output is the adjusted commentary information displayed on the user's device, allowing the user to enjoy a highly personalized viewing experience.

[1421] 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.

[1422] 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.

[1423] 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.

[1424] 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.

[1425] 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.

[1426] 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.

[1427] 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).

[1428] 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.

[1429] 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."

[1430] 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.

[1431] 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).

[1432] 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.

[1433] 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.

[1434] 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.

[1435] 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.

[1436] 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.

[1437] 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.

[1438] 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.

[1439] 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.

[1440] 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.

[1441] 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.

[1442] The following is further disclosed regarding the above embodiment.

[1443] (Claim 1)

[1444] A means of receiving and analyzing game footage in real time,

[1445] A means for extracting information on a specific target player based on the analysis results;

[1446] means for generating commentary on the specific target player based on the extracted information;

[1447] The system includes means for transmitting the generated commentary to a user terminal.

[1448] (Claim 2)

[1449] A means for a user to select a specific target player;

[1450] means for transmitting the selected information to a server;

[1451] 10. The system of claim 1, further comprising: means for receiving and displaying commentary from the server.

[1452] (Claim 3)

[1453] means for generating a natural language commentary based on the extracted information;

[1454] 3. The system of claim 1, further comprising: means for converting the generated natural language commentary into audio data.

[1455] "Example 1"

[1456] (Claim 1)

[1457] A means of receiving and analyzing game footage in real time,

[1458] means for identifying a specific target player based on the analysis results;

[1459] means for extracting information about the identified target player;

[1460] A means for generating a commentary in natural language based on the extracted information;

[1461] means for converting the generated natural language commentary into audio data;

[1462] The system includes means for transmitting the generated commentary to a user terminal.

[1463] (Claim 2)

[1464] A means for a user to select a specific target player;

[1465] means for transmitting the selected information to a server;

[1466] means for generating a request including identification information of the selected specific target player;

[1467] means for transmitting the generated request to a server;

[1468] 10. The system of claim 1, further comprising: means for receiving and displaying commentary from the server.

[1469] (Claim 3)

[1470] A means for generating commentary using a generative AI model to generate information in natural language;

[1471] means for converting the generated natural language commentary into audio data;

[1472] 10. The system of claim 1, further comprising: means for transmitting the generated commentary to a user terminal in real time.

[1473] "Application Example 1"

[1474] (Claim 1)

[1475] A means of receiving and analyzing game footage in real time,

[1476] A means for extracting information on a specific target player based on the analysis results;

[1477] means for generating commentary on the specific target player based on the extracted information;

[1478] means for transmitting the generated commentary to a user terminal;

[1479] The system includes means for displaying and audibly reproducing the received commentary at a user terminal.

[1480] (Claim 2)

[1481] A means for a user to select a specific target player;

[1482] means for transmitting the selected information to a server;

[1483] 10. The system of claim 1, further comprising means for receiving and displaying commentary from the server.

[1484] (Claim 3)

[1485] means for generating a natural language commentary based on the extracted information;

[1486] 3. The system of claim 1, further comprising: means for converting the generated natural language commentary into audio data.

[1487] "Example 2: Combining Emotion Engines"

[1488] (Claim 1)

[1489] A means of receiving and analyzing game footage in real time,

[1490] A means for extracting information on a specific target player based on the analysis results;

[1491] means for generating commentary on the specific target player based on the extracted information;

[1492] means for recognizing the emotional state of a user;

[1493] means for dynamically adjusting commentary content based on the recognized emotion information;

[1494] The system includes means for transmitting the generated commentary to a user terminal.

[1495] (Claim 2)

[1496] A means for a user to select a specific target player;

[1497] means for transmitting the selected information to a server;

[1498] 10. The system of claim 1, further comprising: means for receiving and displaying commentary from the server.

[1499] (Claim 3)

[1500] means for generating a natural language commentary based on the extracted information;

[1501] and means for converting the generated natural language commentary into audio data.

[1502] "Application example 2 when combining emotion engines"

[1503] (Claim 1)

[1504] A means of receiving and analyzing game footage in real time,

[1505] A means for extracting information on a specific target player based on the analysis results;

[1506] means for generating commentary on the specific target player based on the extracted information;

[1507] means for transmitting the generated commentary to a user terminal;

[1508] means for recognizing and analyzing the emotional state of a user;

[1509] means for dynamically adjusting commentary content based on the user's emotional state;

[1510] A system including:

[1511] (Claim 2)

[1512] A means for a user to select a specific target player;

[1513] means for transmitting the selected information to a server;

[1514] 10. The system of claim 1, further comprising: means for receiving and displaying commentary from the server.

[1515] (Claim 3)

[1516] means for generating a natural language commentary based on the extracted information;

[1517] and means for converting the generated natural language commentary into audio data. [Explanation of symbols]

[1518] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving and analyzing game footage in real time, A means for extracting information on a specific target player based on the analysis results; means for generating commentary on the specific target player based on the extracted information; The system includes means for transmitting the generated commentary to a user terminal.

2. A means for a user to select a specific target player; means for transmitting the selected information to a server; 2. The system of claim 1, further comprising: means for receiving and displaying commentary from a server.

3. means for generating a natural language commentary based on the extracted information; 3. The system of claim 1, further comprising: means for converting the generated natural language commentary into audio data.

Images