System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving video data of a sporting event from a user terminal; means for analyzing the received video data to identify a particular sporting event; means for generating live commentary, commentary, and cheering audio for the identified sporting event; means for combining the generated live commentary, commentary, and cheering audio with the video data; and means for providing the combined video data to the 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] Interest in sports has grown in recent years, but professional commentary, narration, and cheering are needed to enhance the viewing experience. However, videos of amateur, amateur baseball, and children's games shot by ordinary users lack a sense of realism and do not fully realize their value. Furthermore, recording games and creating game reports requires a great deal of effort, making it difficult to enjoy them casually. Therefore, there is a need for a system that can easily add professional-level commentary and narration and generate game articles and game reports. [Means for solving the problem]

[0005] The present invention provides a means for receiving video data of a sporting event from a user terminal and analyzing the video data to identify specific sporting events. This identification means can identify events such as scoring scenes and important plays. The system also includes a means for generating play-by-play, commentary, and cheering audio corresponding to the identified sporting event, and a means for synthesizing the generated audio with the video data. Furthermore, the system includes a means for automatically generating match articles and spectator reports based on information about the identified sporting event, thereby achieving a consistently professional viewing experience and enabling users to easily create high-quality sports videos and records.

[0006] A "user terminal" is a device operated by a user, and refers to digital devices such as smartphones, tablets, and personal computers.

[0007] "Video data" refers to video data of filmed sports events, and refers to digital files containing visual information.

[0008] "Server" refers to a main machine that receives and analyzes and processes data sent from a user terminal, or a computing device that provides that function.

[0009] "Analysis" refers to the process of using computer technology to analyze the content contained in video data and identify specific events or patterns.

[0010] A "sporting event" is a specific action or occurrence that occurs during a sporting competition, including goals, shots, points, fouls, etc.

[0011] "Live commentary" refers to audio information that provides real-time commentary on the progress of a sports competition and the movements of players.

[0012] "Commentary" refers to audio information that provides expert information and opinions regarding the playing situation and the play of players.

[0013] "Cheer sounds" refers to the sounds of cheers and support from spectators, and includes sound effects added to enhance the sense of realism of the match.

[0014] "Synthesis" refers to the process of combining different audio and video data into a single whole.

[0015] "Match reports" refer to text information describing the results and important scenes of sports competitions.

[0016] A "match report" refers to text information that includes details and highlights of the match, as well as impressions and analysis from a spectator's perspective. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This invention is a form for realizing a system that uses video data of sports events filmed by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0039] User device:

[0040] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0041] server:

[0042] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0043] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form via a speech generation module.

[0044] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0045] Additionally, the server automatically generates match articles and match reports based on the identified sporting event information, similarly using natural language generation AI to create detailed descriptions of the match's highlights and key moments.

[0046] Examples:

[0047] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0048] This allows users to receive professional commentary and commentary along with detailed match coverage, providing a great sports experience.

[0049] The processing flow will be explained below.

[0050] Step 1:

[0051] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0052] Step 2:

[0053] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0054] Step 3:

[0055] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0056] Step 4:

[0057] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0058] Step 5:

[0059] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0060] Step 6:

[0061] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0062] Step 7:

[0063] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0064] Step 8:

[0065] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0066] Step 9:

[0067] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0068] Step 10:

[0069] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0070] Step 11:

[0071] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0072] Example 1

[0073] 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."

[0074] In recent years, there has been an increase in opportunities to film sporting events using digital devices such as smartphones and tablets. However, the filmed videos often lack a sense of realism for viewers because they are not accompanied by professional commentary, narration, or cheering sounds. Furthermore, creating match articles or spectator reports based on filmed footage requires a great deal of time and effort. To solve this problem, a system is needed that can automatically generate commentary, narration, and cheering sounds from filmed video data and synthesize them into the video data.

[0075] 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.

[0076] In this invention, the server includes means for receiving video data of a sports event from a user terminal, means for analyzing the received video data using computer vision technology and a pattern recognition algorithm to identify a specific sports event, means for generating commentary, narration, and cheering sounds for the identified sports event using natural language generation, means for converting the generated commentary, narration, and cheering sounds into an audio file using an audio generation system and synthesizing the audio with the video data, means for providing the synthesized video data to the user terminal, means for automatically generating a match article and a spectator report using natural language generation based on information about the identified sports event, and means for providing the generated match article and spectator report to the user terminal. This makes it possible to automatically add professional commentary, narration, and cheering sounds to filmed sports video, and further generate a match article and a spectator report.

[0077] A "user terminal" is a digital device used by a user, such as a smartphone, tablet, or PC.

[0078] "Video data" is digital data containing moving images of a sporting event.

[0079] A "server" is a computer system that analyzes, processes, and stores data received from user terminals.

[0080] "Computer vision technology" is a technology that allows computers to analyze images and videos and understand their meaning.

[0081] A "pattern recognition algorithm" is a mathematical technique for identifying specific patterns or features in data.

[0082] "Sporting event" refers to a specific action or occurrence in a sporting competition (e.g., a goal, a foul).

[0083] "Natural language generation" is a technology that allows a computer to automatically create natural-sounding sentences based on data.

[0084] "Commentary" refers to audio or text that verbally describes the progress of a sporting event or important events.

[0085] "Cheering sounds" are sounds that imitate the cheers of spectators at a sporting event.

[0086] A "voice generation system" is a technology for converting text data into voice data.

[0087] An "audio file" is audio data stored in digital format.

[0088] A "video editing module" is software for integrating multiple audio and video data into a single viewable format.

[0089] A "match report" is a written report of the results and highlights of a sporting event.

[0090] A "game report" is a piece of writing in which a spectator gives a detailed description of their impressions of the game and important events that occurred.

[0091] This invention is a system that uses video data of sports events taken by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0092] User terminal

[0093] User devices are digital devices such as smartphones, tablets, and PCs. Users use a dedicated application or web interface to upload video data of sports events they have filmed to a server. For example, a user selects a video of their child's soccer game in the dedicated application and presses the upload button, sending the data to the server.

[0094] server

[0095] The server analyzes the video data received from the user device using computer vision techniques (e.g., OpenCV, TENSORFLOW®) and pattern recognition algorithms (e.g., YOLO, SSD). The server identifies specific sporting events (e.g., goals and fouls) in each video frame and records the timestamps of those events.

[0096] Once the analysis is complete, the server generates commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4 (registered trademark)) is used to create realistic commentary and commentary. For example, if a goal is detected, a commentary text such as "Player A scores a great goal!" is generated. The generated text is then converted into an audio file using a speech generation system (e.g., WaveNet, Tacotron 2).

[0097] The server then uses a video editing module (e.g., FFmpeg, Adobe Premiere Pro) to combine the generated audio files with the original video data. The commentary, commentary, and cheering audio are placed appropriately based on the video timestamps, resulting in a professional-looking result. The final combined video data is then saved.

[0098] Furthermore, the server uses natural language generation AI (e.g., GPT-4) to create match articles and match reports based on the identified sporting event information, such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0099] This allows users to receive professional commentary and commentary along with detailed match coverage.

[0100] Specific examples

[0101] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (e.g., goals scored or fouls scored). When a goal is detected, a corresponding commentary such as "What a great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as, "Player A scored a great goal in the fifth minute of the first half, deciding the game."

[0102] Prompt Sentence Examples

[0103] "A user has uploaded a video of their child's soccer match. Please analyze the goal and foul scenes, generate commentary, narration, and cheering sounds based on that, and synthesize them into the video. You will also need to create a match article that details the highlights and important moments of the match."

[0104] This system allows users to easily obtain professional-quality sports footage and match reports, allowing them to enjoy a more immersive sports experience.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] The user uploads video data to the server. Using a smartphone, tablet, or PC, the user opens a dedicated application or web interface. Next, the user selects the video data of a sports event using the file selection function and operates the upload button to send the data to the server. The input here is the video data shot by the user, and the output is the data uploaded to the server.

[0108] Step 2:

[0109] The server analyzes the received video data. The server uses computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD) to analyze the video data frame by frame. Specifically, the server divides the video data into frames, identifies specific sporting events (e.g., goal scenes or foul scenes) in each frame, and records the timestamps of those events. The input is the video data uploaded to the server, and the output is a list of specific sporting events and their timestamps.

[0110] Step 3:

[0111] The server generates commentary, narration, and cheering audio. Based on the identified sporting events, the server uses natural language generation AI (e.g., GPT-4) to generate commentary, narration, and cheering audio text. For example, if a goal is detected, the server generates commentary text such as "Player A scores a great goal!". The server then uses a speech generation system (e.g., WaveNet, Tacotron 2) to convert this text into an audio file. The input is a list of specific sporting events and their timestamps, and the output is the generated audio files of commentary, narration, and cheering audio.

[0112] Step 4:

[0113] The server synthesizes the audio into the video. Using a video editing module (e.g., FFmpeg, Adobe Premiere Pro), the server synthesizes the generated audio file with the original video data. Specifically, the server appropriately positions the generated commentary, narration, and cheering audio based on the corresponding timestamps in the video, providing a realistic sense of presence. For example, a commentary saying "Great shot!" is inserted at the right time just before a goal is scored. The input is the original video data and the generated audio file, and the output is the final synthesized video data.

[0114] Step 5:

[0115] The server generates match articles and viewing reports. Based on the information of the identified sporting event, the server uses natural language generation AI (e.g., GPT-4) to create match articles and viewing reports. For example, it generates a sentence such as, "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game." The input is information about a specific sporting event, and the output is the generated match articles and viewing reports.

[0116] Step 6:

[0117] The server provides the generated match articles and observation reports to the user terminal. The server sends the synthesized video data and the generated match articles and observation reports to the user terminal. The user can view and enjoy this data using the terminal. The input is the synthesized video data, match articles, and observation reports, and the output is the final data provided to the user terminal.

[0118] (Application example 1)

[0119] 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."

[0120] Conventional sports video simply provides filmed footage, lacking a sense of realism due to the lack of commentary, commentary, or cheering sounds. It also faces the problem of being unable to identify important scenes and events in a game and provide appropriate commentary or articles in real time. Furthermore, there is a lack of easy ways to share the generated content on social media, etc. This makes it difficult to fully bring out the appeal of sports video shot by users.

[0121] 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.

[0122] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data to identify a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, and means for sharing the generated video data, match reports, and spectator reports on external platforms such as social media. This allows users to enjoy video with professional commentary and commentary synthesized, and also makes it easy to share the generated match reports and spectator reports.

[0123] A "user terminal" is an electronic device operated by a user, and includes smartphones, tablets, personal computers, and the like.

[0124] "Video data" refers to digital data containing visual information of a sports event captured by a user.

[0125] "Analysis" refers to the process of identifying specific sporting events using computer vision technology and pattern recognition algorithms from the received video data.

[0126] A "particular sporting event" is a significant moment in the flow of a sporting event, such as a goal, a foul, or a significant play.

[0127] "Commentary" refers to the real-time verbal description and commentary of a particular sporting event.

[0128] "Commentary" means explanations that provide additional information and background about the progress or play of a sporting event.

[0129] "Cheering sounds" refers to the generation and playback of sounds that express the excitement of a game, such as the cheers and applause of spectators.

[0130] "Synthesis" refers to the process of integrating the generated commentary, commentary, and cheering sounds into the original video data to create an integrated video and audio.

[0131] A "match report" is a piece of text that provides detailed descriptions of important moments and results of a match.

[0132] A "match report" is a piece of writing that includes the overall flow and highlights of the match, as well as the spectator's impressions and opinions.

[0133] "Social media" refers to online platforms for sharing and interacting with user-generated content.

[0134] "External Platform" refers to other online services or websites that are different from this system but can be linked to it.

[0135] This invention provides a system that automatically generates commentary, commentary, and cheering sounds from video data of sports events taken by users, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module.

[0136] User terminal

[0137] The user devices are digital devices such as smartphones, tablets, and PCs, and users use them to capture video data of sports events and upload it to a server via a dedicated application or web interface. The video data captured by users is stored in cloud storage (e.g., Amazon S3).

[0138] server

[0139] The server has the function of analyzing the video data stored in the cloud storage. First, it analyzes the received video data using computer vision technology (e.g., OpenCV) to identify specific sporting events (goals, fouls, important plays, etc.). During this analysis process, timestamps of specific events are recorded.

[0140] Next, based on the identified events, a generative AI model (e.g., GPT-4) is used to generate commentary, commentary, and cheers. The generated text is output as an audio file via a speech synthesis API (e.g., Google® Cloud Text-to-Speech).

[0141] The generated commentary, commentary, and cheering audio is then combined with the original video data using the API of video editing software (e.g., Adobe Premiere Pro) to create professional video content.

[0142] The server also uses the generative AI model to generate match articles and game reports based on the analysis results. These reports are delivered to users' devices and can be shared on external platforms such as social media.

[0143] Specific examples

[0144] For example, a user films a child's soccer game and uploads the footage to a server via a smartphone application. The server analyzes the video data and identifies specific events (e.g., a goal). For the identified event, a generative AI model generates commentary text such as "Great shot!" and a voice synthesis API is used to create an audio file. The generated audio is then combined with the original video, creating immersive video content. The generative AI model is also used to create match articles and spectator reports, providing text such as "Player A scored a brilliant goal in the fifth minute of the first half, sealing the victory."

[0145] Prompt Sentence Examples

[0146] 1. Video analysis prompt:

[0147] "Please identify the important sporting events (goals, fouls, etc.) in this video."

[0148] 2. Commentary generation prompts:

[0149] "Generate commentary text for goals scored in a match. For example, 'Great shot!'"

[0150] 3. Match article generation prompt:

[0151] "Generate an article detailing the highlights and key moments of the match. For example, 'Player A scored a brilliant goal in the fifth minute of the first half, deciding the game.'"

[0152] As described above, this system enables users to easily create and share professional sports video content.

[0153] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0154] Step 1:

[0155] Input: Video data of sports events taken by the user

[0156] How it works: Users use their smartphone, tablet, or PC to upload footage of sports events to a server via a dedicated application or web interface.

[0157] Output: Video data stored in cloud storage (e.g. Amazon S3)

[0158] Step 2:

[0159] Input: Video data stored in cloud storage

[0160] How it works: The server retrieves video data stored in cloud storage and analyzes the video using computer vision techniques (e.g., OpenCV). This analysis identifies specific sporting events (e.g., goals, fouls, key plays) and captures event timestamps.

[0161] Output: timestamp and information about identified sporting events

[0162] Step 3:

[0163] Input: Timestamp and information of identified sporting events

[0164] How it works: The server uses a generative AI model (e.g., GPT-4) to generate commentary, commentary, and cheers for the identified events. Depending on the text you want to generate, input the following prompts into the generative AI model:

[0165] Example prompt: "Generate commentary text for a goal in a game. For example, 'Great shot!'"

[0166] Output: Text of the generated commentary, commentary, and cheers

[0167] Step 4:

[0168] Input: Generated commentary, commentary, and cheering audio text

[0169] How it works: The server uses a speech synthesis API (e.g., Google Cloud Text-to-Speech) to output the generated commentary, commentary, and cheering text as an audio file.

[0170] Output: Generated audio file

[0171] Step 5:

[0172] Input: Video data and generated audio files

[0173] How it works: The server uses the API of video editing software (e.g. Adobe Premiere Pro) to synthesize the generated commentary, commentary, and cheering audio into the original video data.

[0174] Output: Composite video data

[0175] Step 6:

[0176] Input: Timestamp and information of identified sporting events

[0177] How it works: The server uses the generative AI model to generate match articles and game reports. Based on the analysis results, it inputs the following prompt sentences into the generative AI model:

[0178] Example prompt: "Generate an article detailing the highlights and key moments of a match. For example, 'Player A scored a spectacular goal in the fifth minute of the first half, deciding the game.'"

[0179] Output: Generated match articles and match reports

[0180] Step 7:

[0181] Input: Synthesized video data and generated match articles and reports

[0182] Operation: The server provides the synthesized video data and generated match articles and spectator reports to the user's device, as well as a means to share them on external platforms such as social media.

[0183] Output: Composite images and articles provided to the user's device, content shared on external platforms

[0184] This allows users to enjoy video content in which professional commentary, commentary, and cheering sounds are combined with footage of sports events they have filmed themselves, and they can also easily share this content with other people.

[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 combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds from video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0187] User device:

[0188] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0189] server:

[0190] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0191] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[0192] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0193] Emotion Engine:

[0194] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0195] Automatic match coverage and match reports:

[0196] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0197] Examples:

[0198] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0199] The system allows users to receive professional commentary and commentary, detailed match coverage, and enjoy a more personalized viewing experience based on their emotions.

[0200] The processing flow will be explained below.

[0201] Step 1:

[0202] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0203] Step 2:

[0204] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0205] Step 3:

[0206] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0207] Step 4:

[0208] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0209] Step 5:

[0210] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0211] Step 6:

[0212] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0213] Step 7:

[0214] The server uses an emotion engine to recognize the user's emotional state, analyzing the user's facial expressions, tone of voice, touch input, and other data acquired from the device to identify emotions.

[0215] Step 8:

[0216] The server adjusts the commentary, commentary, and cheering sounds it generates based on the user's perceived emotional state, for example, increasing the tone of the commentary if the user is perceived to be happy.

[0217] Step 9:

[0218] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0219] Step 10:

[0220] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0221] Step 11:

[0222] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0223] Step 12:

[0224] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0225] Step 13:

[0226] The server also applies the user's emotional information to match articles and game reports, generating sentences that emphasize scenes that the user found particularly exciting, for example.

[0227] Step 14:

[0228] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0229] Example 2

[0230] 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."

[0231] Conventional video analysis systems for sports events have been unable to take into account the user's emotional state when automatically generating and synthesizing commentary, narration, and cheering sounds from video data, making it difficult to provide a personalized viewing experience.In addition, there has been a lack of easy ways for users to obtain detailed match articles and spectator reports based on the results of video data analysis.

[0232] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0233] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, means for recognizing a user's emotion, means for adjusting the generated commentary and commentary based on the user's emotional state, and means for providing the synthesized video data to the user terminal. This makes it possible to provide a more personalized viewing experience based on the user's emotion. Furthermore, by adding a function for automatically generating match articles and viewing reports, users can easily obtain detailed analysis information.

[0234] A "user terminal" is a device operated by a user, and refers to digital devices such as smartphones, tablets, and personal computers.

[0235] "Video data" refers to recorded video of a sports event transmitted from a user terminal.

[0236] "Server" refers to a computer system that analyzes video data received from user terminals and performs processing, generation, and synthesis.

[0237] "Analysis" refers to the process of analyzing video data to identify specific sporting events.

[0238] A "sporting event" refers to a specific occurrence during a sporting competition (e.g., a goal or a foul).

[0239] "Identification" refers to finding specific sporting events from the analyzed data.

[0240] "Live commentary" refers to the content of a sports event being explained verbally in real time.

[0241] "Explanation" refers to supplementary information such as technical aspects and background information about sports competitions.

[0242] "Cheer sounds" refers to the sounds of cheers inserted into exciting scenes such as sporting events.

[0243] "Generated" refers to the use of AI technology and algorithms to create commentary, commentary, and cheering audio.

[0244] "Synthesis" refers to the process of integrating generated commentary, commentary, and cheering sounds into the original video data.

[0245] "Emotion recognition" refers to technology that analyzes a user's emotional state from facial expressions, tone of voice, and touch input.

[0246] "Adjusting" refers to changing the tone and content of the commentary or commentary generated based on the user's emotional state.

[0247] A "match report" refers to an article that describes important scenes or events in a sports competition in written form.

[0248] "Spectator report" refers to a report that includes detailed analytical information about a sports competition.

[0249] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds for video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0250] User terminal

[0251] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0252] server

[0253] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques and pattern recognition algorithms (e.g., OpenCV, TensorFlow), and records the event timestamps.

[0254] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to generate realistic commentary and commentary. The generated audio is then converted to a realistic form using a speech generation module (e.g., Amazon Polly, Google Text-to-Speech).

[0255] The server then uses a video editing module (e.g., FFmpeg) to properly place commentary, commentary, and cheering audio into the video, enhancing the sense of realism.

[0256] Emotion Engine

[0257] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. This recognition uses facial expression analysis technology (e.g., dlib), voice tone analysis technology, and touch input analysis technology. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0258] Automatic generation of match articles and match reports

[0259] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0260] Specific examples

[0261] For example, suppose a user films a child's soccer game and uploads the video. The server analyzes the video data and detects specific events such as goals and fouls. When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narration will become more excited.

[0262] Furthermore, based on the analysis results of the entire match, match articles such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game" are automatically generated.

[0263] Prompt Sentence Examples

[0264] Examples of prompts to be input to a generative AI model include:

[0265] "We analyzed a video of a children's soccer game. There was a goal. Please generate audio commentary such as 'Great shot!' and cheers for the goal. Also, since the user's emotions are very happy, please adjust the tone of the commentary to be more excited."

[0266] The system allows users to receive professional commentary and commentary, detailed match coverage, and a personalized viewing experience tailored to their emotions.

[0267] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0268] Step 1:

[0269] Upload video data from the user's device

[0270] Description: Users access a dedicated application or web interface using their own devices such as smartphones or PCs. After selecting the video data and pressing the upload button, the video data is sent to the server.

[0271] Input: Video data file

[0272] Output: Notification of completion of video data transmission to the server

[0273] Specific operation: The user taps the "Upload Video" button in the app, selects a video file from the device's storage, and then presses the "Upload" button to send the file to the server.

[0274] Step 2:

[0275] Video data analysis by the server

[0276] Description: The server analyzes the received video data using computer vision techniques and pattern recognition algorithms to identify specific sporting events (e.g. goals and fouls) and record them with timestamps.

[0277] Input: Uploaded video data

[0278] Output: Sport event identification results and timestamps

[0279] How it works: Once the video data arrives at the server, the video analysis process begins automatically. The server analyzes each frame, tracking player movements and ball position to detect events like goals and fouls. Detected events are then stored in a database with a timestamp.

[0280] Step 3:

[0281] Server-based generation of commentary, commentary, and cheering sounds

[0282] Description: Based on the identified event information, the server generates corresponding commentary, commentary, and cheering audio using a generative AI model (e.g., GPT-4), and uses speech synthesis technology (e.g., Amazon Polly, Google Text-to-Speech) to create an audio file from the generated text.

[0283] Input: Sport event identification results, timestamp

[0284] Output: Translated commentary and cheers text and audio files

[0285] Specific operation: For each event, the server inputs the following prompt sentence into the AI ​​model: "A goal scene has been detected. Please generate commentary text that says, 'Great shot!'" The generated text is then passed to a speech synthesis library to create an audio file.

[0286] Step 4:

[0287] Server-based audio and video data synthesis

[0288] Description: The server synthesizes the generated audio file into the video data. Using a video editing module (e.g., FFmpeg), it places the commentary, commentary, and cheering audio appropriately into the video.

[0289] Input: Original video data, generated audio file, timestamp

[0290] Output: Composite video file

[0291] How it works: Once the speech synthesis is complete, the server uses FFmpeg to add the audio file to the original video based on the timestamp. For example, a commentary saying "Great shot!" could be placed at the timestamp of a goal, creating a more immersive video.

[0292] Step 5:

[0293] Emotion recognition and feedback for users using an emotion engine

[0294] Description: The emotion engine analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state, and adjusts the tone of the commentary or commentary it generates based on the recognized emotional state.

[0295] Input: User facial expression data, tone of voice data, touch input data

[0296] Output: User's emotional state, adjusted commentary text

[0297] How it works: The emotion engine analyzes the user's previous viewing behavior and feedback. For example, if the user is very happy, the next commentary generated will be set to a higher level of excitement.

[0298] Step 6:

[0299] Automatic generation of match articles and match reports

[0300] Description: The server uses natural language generation AI to automatically generate match articles and match reports based on the analysis results and identified event information.

[0301] Input: Sporting event identification results

[0302] Output: Automatically generated match articles and match reports

[0303] What it does: The server inputs details of a sporting event into the AI ​​model and generates "match highlights and articles," such as "Player A scored a spectacular goal in the fifth minute of the first half, deciding the game."

[0304] Step 7:

[0305] Delivery of results to user devices

[0306] Description: The server delivers the generated video and match reports to the user's device, allowing the user to enjoy professional commentary, commentary, and detailed reports.

[0307] Input: Composite video files, automatically generated match reports and game reports

[0308] Output: Notification and download link to user's device

[0309] What happens: The server sends the created video file and article to the user via the app or web interface. The user logs in to download or watch the content. The app displays a notification saying "New video available," which the user can tap to go to the viewing screen.

[0310] The system provides users with a personalized and professional viewing experience.

[0311] (Application example 2)

[0312] 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."

[0313] Conventional sports viewing systems lack the means to automatically generate and synthesize realistic commentary, commentary, and cheering sounds for user-recorded video, resulting in a less than satisfactory viewing experience. It is also difficult to provide personalized content that reflects the user's emotions. This can result in a decrease in the viewer's sense of immersion and satisfaction.

[0314] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, narration, and cheering sounds for the identified sporting event, means for recognizing the user's emotion and adjusting the commentary, narration, and cheering sounds based on the emotion, means for synthesizing the generated commentary, narration, and cheering sounds with the video data, and means for providing the synthesized video data to the user terminal. This allows the user to enjoy video with professional commentary and narration, and a viewing experience personalized according to their emotion.

[0315] "User terminal" refers to digital devices such as smartphones, tablets, and personal computers operated by users.

[0316] "Video Data" refers to data in digital format that includes footage of a sports competition.

[0317] A "sporting event" refers to a specific significant moment or occurrence in a sporting competition, such as a goal or a foul.

[0318] "Live commentary" refers to audio commentary that explains and reports the situation and plays during a sports competition in real time.

[0319] "Commentary" refers to audio commentary that includes detailed explanations of the content, tactics, and player movements of a sports competition.

[0320] "Cheer sounds" refers to sounds that simulate the cheers of spectators, expressing the excitement and thrill of moments at sporting events.

[0321] "Means for recognizing emotions" refers to a system that analyzes and identifies a user's emotional state from facial images, tone of voice, touch input, etc.

[0322] "Means for adjusting commentary, commentary, and cheering sounds" refers to a system that changes the content and tone of the commentary, commentary, and cheering sounds generated based on the user's emotional state.

[0323] "Means for synthesis" refers to the technology for overlaying the generated commentary, commentary and cheering sounds onto the original video data.

[0324] This invention is a system that automatically generates and synthesizes commentary, narration, and cheering sounds for video data of sports events by operating multiple modules in coordination, providing users with a personalized viewing experience. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes an emotion engine.

[0325] User terminal

[0326] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. The user selects the video data using a dedicated application and operates the upload button, which sends the data to the server.

[0327] server

[0328] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques (e.g., OpenCV) and pattern recognition algorithms, and records the event timestamps.

[0329] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT model) is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[0330] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0331] Emotion Engine

[0332] The emotion engine is a module for recognizing the user's emotions. It analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state. Based on the emotional state recognized by the emotion engine, the server adjusts the generated commentary, narration, and cheering audio accordingly. This process uses the Emotion Recognizer library (e.g., EmotionRecognizer).

[0333] Automatic generation of match articles and match reports

[0334] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used to generate text that describes the highlights and important scenes of the match in detail. The generated match articles and spectator reports are also provided to the user's device.

[0335] Specific examples

[0336] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0337] Prompt Sentence Examples

[0338] An example of a prompt for a generative AI model is, "The user is very excited. Please commentate by saying, 'Great shot!' when a goal is scored."

[0339] This allows users to enjoy professional commentary and commentary, as well as a personalized viewing experience tailored to their emotions.

[0340] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0341] Step 1:

[0342] Uploading video data from user devices

[0343] Users use digital devices such as smartphones, tablets, and PCs to upload video data of sports events they have taken to a server via a dedicated application. The input is the video file taken by the user, and the output is the video file being transferred to the server.

[0344] Step 2:

[0345] Video data analysis

[0346] The server analyzes the received video data and identifies specific sporting events. This analysis uses computer vision techniques such as OpenCV and pattern recognition algorithms. The input is the uploaded video file, and the output is the timestamp information of the identified sporting events. Specifically, it sequentially scans each frame of the video data to detect characteristic events such as goals and fouls.

[0347] Step 3:

[0348] User Emotion Recognition

[0349] The server analyzes the user's facial expressions and tone of voice to recognize the user's emotions. The EmotionRecognizer library is used here. The input is the user's facial image and voice data, and the output is the recognized emotional state (e.g., joy, excitement). Specifically, it extracts features from the user's facial image and inputs them into a model to estimate emotions.

[0350] Step 4:

[0351] Generation of commentary, commentary and cheering sounds

[0352] The server generates commentary and cheering audio based on the identified sporting event and the user's emotional state. This process utilizes natural language generation AI (e.g., GPT model). The input is the timestamp information of the sporting event and the user's emotional state, and the output is a specific audio file. Specifically, the template sentence corresponding to the identified event is used as the prompt sentence, and the generative AI model generates natural commentary and commentary, which is then converted into audio.

[0353] Step 5:

[0354] Synthesizing live commentary, commentary, and cheering sounds into video data

[0355] The server synthesizes the generated commentary, commentary, and cheering audio into the original video data. This is done using a video editing module. The input is the original video file and the generated audio file, and the output is the synthesized video data. Specifically, the audio file is inserted and synchronized at the appropriate position in the video file according to the timestamp.

[0356] Step 6:

[0357] Providing the synthesized video data to the user's device

[0358] The server distributes the synthesized video data to the user terminal. The input is the synthesized video file, and the output is a video file encoded in a format that can be played on the user terminal. Specifically, the video data is provided to the user terminal in streaming or download format.

[0359] Step 7:

[0360] Automatic generation of match articles and match reports

[0361] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used here. The input is the timestamp information and related detailed data of the sporting event, and the output is the generated text report. Specifically, it generates prompt sentences based on the highlights of each event, and then uses the natural language generation AI to create detailed sentences.

[0362] Prompt Sentence Examples

[0363] The prompt sentence is, "The user is very excited. Please commentate when the goal is scored, saying, 'Great shot!'"

[0364] 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.

[0365] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0366] 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.

[0367] [Second embodiment]

[0368] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0369] 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.

[0370] 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).

[0371] 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.

[0372] 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.

[0373] 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).

[0374] 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.

[0375] 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.

[0376] 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.

[0377] 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.

[0378] 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.

[0379] 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."

[0380] This invention is a form for realizing a system that uses video data of sports events filmed by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0381] User device:

[0382] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0383] server:

[0384] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0385] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form via a speech generation module.

[0386] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0387] Additionally, the server automatically generates match articles and match reports based on the identified sporting event information, similarly using natural language generation AI to create detailed descriptions of the match's highlights and key moments.

[0388] Examples:

[0389] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0390] This allows users to receive professional commentary and commentary along with detailed match coverage, providing a great sports experience.

[0391] The processing flow will be explained below.

[0392] Step 1:

[0393] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0394] Step 2:

[0395] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0396] Step 3:

[0397] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0398] Step 4:

[0399] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0400] Step 5:

[0401] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0402] Step 6:

[0403] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0404] Step 7:

[0405] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0406] Step 8:

[0407] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0408] Step 9:

[0409] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0410] Step 10:

[0411] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0412] Step 11:

[0413] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0414] Example 1

[0415] 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."

[0416] In recent years, there has been an increase in opportunities to film sporting events using digital devices such as smartphones and tablets. However, the filmed videos often lack a sense of realism for viewers because they are not accompanied by professional commentary, narration, or cheering sounds. Furthermore, creating match articles or spectator reports based on filmed footage requires a great deal of time and effort. To solve this problem, a system is needed that can automatically generate commentary, narration, and cheering sounds from filmed video data and synthesize them into the video data.

[0417] 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.

[0418] In this invention, the server includes means for receiving video data of a sports event from a user terminal, means for analyzing the received video data using computer vision technology and a pattern recognition algorithm to identify a specific sports event, means for generating commentary, narration, and cheering sounds for the identified sports event using natural language generation, means for converting the generated commentary, narration, and cheering sounds into an audio file using an audio generation system and synthesizing the audio with the video data, means for providing the synthesized video data to the user terminal, means for automatically generating a match article and a spectator report using natural language generation based on information about the identified sports event, and means for providing the generated match article and spectator report to the user terminal. This makes it possible to automatically add professional commentary, narration, and cheering sounds to filmed sports video, and further generate a match article and a spectator report.

[0419] A "user terminal" is a digital device used by a user, such as a smartphone, tablet, or PC.

[0420] "Video data" is digital data containing moving images of a sporting event.

[0421] A "server" is a computer system that analyzes, processes, and stores data received from user terminals.

[0422] "Computer vision technology" is a technology that allows computers to analyze images and videos and understand their meaning.

[0423] A "pattern recognition algorithm" is a mathematical technique for identifying specific patterns or features in data.

[0424] "Sporting event" refers to a specific action or occurrence in a sporting competition (e.g., a goal, a foul).

[0425] "Natural language generation" is a technology that allows a computer to automatically create natural-sounding sentences based on data.

[0426] "Commentary" refers to audio or text that verbally describes the progress of a sporting event or important events.

[0427] "Cheering sounds" are sounds that imitate the cheers of spectators at a sporting event.

[0428] A "voice generation system" is a technology for converting text data into voice data.

[0429] An "audio file" is audio data stored in digital format.

[0430] A "video editing module" is software for integrating multiple audio and video data into a single viewable format.

[0431] A "match report" is a written report of the results and highlights of a sporting event.

[0432] A "game report" is a piece of writing in which a spectator gives a detailed description of their impressions of the game and important events that occurred.

[0433] This invention is a system that uses video data of sports events taken by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0434] User terminal

[0435] User devices are digital devices such as smartphones, tablets, and PCs. Users use a dedicated application or web interface to upload video data of sports events they have filmed to a server. For example, a user selects a video of their child's soccer game in the dedicated application and presses the upload button, sending the data to the server.

[0436] server

[0437] The server analyzes the video data received from the user device using computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD). The server identifies specific sporting events (e.g., goals and fouls) in each video frame and records the timestamps of those events.

[0438] Once the analysis is complete, the server generates commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to create realistic commentary and commentary. For example, if a goal is detected, commentary text such as "Player A scores a great goal!" is generated. The generated text is then converted into an audio file using a speech generation system (e.g., WaveNet, Tacotron 2).

[0439] The server then uses a video editing module (e.g., FFmpeg, Adobe Premiere Pro) to combine the generated audio files with the original video data. The commentary, commentary, and cheering audio are placed appropriately based on the video timestamps, resulting in a professional-looking result. The final combined video data is then saved.

[0440] Furthermore, the server uses natural language generation AI (e.g., GPT-4) to create match articles and match reports based on the identified sporting event information, such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0441] This allows users to receive professional commentary and commentary along with detailed match coverage.

[0442] Specific examples

[0443] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (e.g., goals scored or fouls scored). When a goal is detected, a corresponding commentary such as "What a great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as, "Player A scored a great goal in the fifth minute of the first half, deciding the game."

[0444] Prompt Sentence Examples

[0445] "A user has uploaded a video of their child's soccer match. Please analyze the goal and foul scenes, generate commentary, narration, and cheering sounds based on that, and synthesize them into the video. You will also need to create a match article that details the highlights and important moments of the match."

[0446] This system allows users to easily obtain professional-quality sports footage and match reports, allowing them to enjoy a more immersive sports experience.

[0447] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0448] Step 1:

[0449] The user uploads video data to the server. Using a smartphone, tablet, or PC, the user opens a dedicated application or web interface. Next, the user selects the video data of a sports event using the file selection function and operates the upload button to send the data to the server. The input here is the video data shot by the user, and the output is the data uploaded to the server.

[0450] Step 2:

[0451] The server analyzes the received video data. The server uses computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD) to analyze the video data frame by frame. Specifically, the server divides the video data into frames, identifies specific sporting events (e.g., goal scenes or foul scenes) in each frame, and records the timestamps of those events. The input is the video data uploaded to the server, and the output is a list of specific sporting events and their timestamps.

[0452] Step 3:

[0453] The server generates commentary, narration, and cheering audio. Based on the identified sporting events, the server uses natural language generation AI (e.g., GPT-4) to generate commentary, narration, and cheering audio text. For example, if a goal is detected, the server generates commentary text such as "Player A scores a great goal!". The server then uses a speech generation system (e.g., WaveNet, Tacotron 2) to convert this text into an audio file. The input is a list of specific sporting events and their timestamps, and the output is the generated audio files of commentary, narration, and cheering audio.

[0454] Step 4:

[0455] The server synthesizes the audio into the video. Using a video editing module (e.g., FFmpeg, Adobe Premiere Pro), the server synthesizes the generated audio file with the original video data. Specifically, the server appropriately positions the generated commentary, narration, and cheering audio based on the corresponding timestamps in the video, providing a realistic sense of presence. For example, a commentary saying "Great shot!" is inserted at the right time just before a goal is scored. The input is the original video data and the generated audio file, and the output is the final synthesized video data.

[0456] Step 5:

[0457] The server generates match articles and viewing reports. Based on the information of the identified sporting event, the server uses natural language generation AI (e.g., GPT-4) to create match articles and viewing reports. For example, it generates a sentence such as, "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game." The input is information about a specific sporting event, and the output is the generated match articles and viewing reports.

[0458] Step 6:

[0459] The server provides the generated match articles and observation reports to the user terminal. The server sends the synthesized video data and the generated match articles and observation reports to the user terminal. The user can view and enjoy this data using the terminal. The input is the synthesized video data, match articles, and observation reports, and the output is the final data provided to the user terminal.

[0460] (Application example 1)

[0461] 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."

[0462] Conventional sports video simply provides filmed footage, lacking a sense of realism due to the lack of commentary, commentary, or cheering sounds. It also faces the problem of being unable to identify important scenes and events in a game and provide appropriate commentary or articles in real time. Furthermore, there is a lack of easy ways to share the generated content on social media, etc. This makes it difficult to fully bring out the appeal of sports video shot by users.

[0463] 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.

[0464] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data to identify a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, and means for sharing the generated video data, match reports, and spectator reports on external platforms such as social media. This allows users to enjoy video with professional commentary and commentary synthesized, and also makes it easy to share the generated match reports and spectator reports.

[0465] A "user terminal" is an electronic device operated by a user, and includes smartphones, tablets, personal computers, and the like.

[0466] "Video data" refers to digital data containing visual information of a sports event captured by a user.

[0467] "Analysis" refers to the process of identifying specific sporting events using computer vision technology and pattern recognition algorithms from the received video data.

[0468] A "particular sporting event" is a significant moment in the flow of a sporting event, such as a goal, a foul, or a significant play.

[0469] "Commentary" refers to the real-time verbal description and commentary of a particular sporting event.

[0470] "Commentary" means explanations that provide additional information and background about the progress or play of a sporting event.

[0471] "Cheering sounds" refers to the generation and playback of sounds that express the excitement of a game, such as the cheers and applause of spectators.

[0472] "Synthesis" refers to the process of integrating the generated commentary, commentary, and cheering sounds into the original video data to create an integrated video and audio.

[0473] A "match report" is a piece of text that provides detailed descriptions of important moments and results of a match.

[0474] A "match report" is a piece of writing that includes the overall flow and highlights of the match, as well as the spectator's impressions and opinions.

[0475] "Social media" refers to online platforms for sharing and interacting with user-generated content.

[0476] "External Platform" refers to other online services or websites that are different from this system but can be linked to it.

[0477] This invention provides a system that automatically generates commentary, commentary, and cheering sounds from video data of sports events taken by users, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module.

[0478] User terminal

[0479] The user devices are digital devices such as smartphones, tablets, and PCs, and users use them to capture video data of sports events and upload it to a server via a dedicated application or web interface. The video data captured by users is stored in cloud storage (e.g., Amazon S3).

[0480] server

[0481] The server has the function of analyzing the video data stored in the cloud storage. First, it analyzes the received video data using computer vision technology (e.g., OpenCV) to identify specific sporting events (goals, fouls, important plays, etc.). During this analysis process, timestamps of specific events are recorded.

[0482] Next, based on the identified events, a generative AI model (e.g., GPT-4) is used to generate commentary, commentary, and cheers, and the generated text is output as an audio file via a speech synthesis API (e.g., Google Cloud Text-to-Speech).

[0483] The generated commentary, commentary, and cheering audio is then combined with the original video data using the API of video editing software (e.g., Adobe Premiere Pro) to create professional video content.

[0484] The server also uses the generative AI model to generate match articles and game reports based on the analysis results. These reports are delivered to users' devices and can be shared on external platforms such as social media.

[0485] Specific examples

[0486] For example, a user films a child's soccer game and uploads the footage to a server via a smartphone application. The server analyzes the video data and identifies specific events (e.g., a goal). For the identified event, a generative AI model generates commentary text such as "Great shot!" and a voice synthesis API is used to create an audio file. The generated audio is then combined with the original video, creating immersive video content. The generative AI model is also used to create match articles and spectator reports, providing text such as "Player A scored a brilliant goal in the fifth minute of the first half, sealing the victory."

[0487] Prompt Sentence Examples

[0488] 1. Video analysis prompt:

[0489] "Please identify the important sporting events (goals, fouls, etc.) in this video."

[0490] 2. Commentary generation prompts:

[0491] "Generate commentary text for goals scored in a match. For example, 'Great shot!'"

[0492] 3. Match article generation prompt:

[0493] "Generate an article detailing the highlights and key moments of the match. For example, 'Player A scored a brilliant goal in the fifth minute of the first half, deciding the game.'"

[0494] As described above, this system enables users to easily create and share professional sports video content.

[0495] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0496] Step 1:

[0497] Input: Video data of sports events taken by the user

[0498] How it works: Users use their smartphone, tablet, or PC to upload footage of sports events to a server via a dedicated application or web interface.

[0499] Output: Video data stored in cloud storage (e.g. Amazon S3)

[0500] Step 2:

[0501] Input: Video data stored in cloud storage

[0502] How it works: The server retrieves video data stored in cloud storage and analyzes the video using computer vision techniques (e.g., OpenCV). This analysis identifies specific sporting events (e.g., goals, fouls, key plays) and captures event timestamps.

[0503] Output: timestamp and information about identified sporting events

[0504] Step 3:

[0505] Input: Timestamp and information of identified sporting events

[0506] How it works: The server uses a generative AI model (e.g., GPT-4) to generate commentary, commentary, and cheers for the identified events. Depending on the text you want to generate, input the following prompts into the generative AI model:

[0507] Example prompt: "Generate commentary text for a goal in a game. For example, 'Great shot!'"

[0508] Output: Text of the generated commentary, commentary, and cheers

[0509] Step 4:

[0510] Input: Generated commentary, commentary, and cheering audio text

[0511] How it works: The server uses a speech synthesis API (e.g., Google Cloud Text-to-Speech) to output the generated commentary, commentary, and cheering text as an audio file.

[0512] Output: Generated audio file

[0513] Step 5:

[0514] Input: Video data and generated audio files

[0515] How it works: The server uses the API of video editing software (e.g. Adobe Premiere Pro) to synthesize the generated commentary, commentary, and cheering audio into the original video data.

[0516] Output: Composite video data

[0517] Step 6:

[0518] Input: Timestamp and information of identified sporting events

[0519] How it works: The server uses the generative AI model to generate match articles and game reports. Based on the analysis results, it inputs the following prompt sentences into the generative AI model:

[0520] Example prompt: "Generate an article detailing the highlights and key moments of a match. For example, 'Player A scored a spectacular goal in the fifth minute of the first half, deciding the game.'"

[0521] Output: Generated match articles and match reports

[0522] Step 7:

[0523] Input: Synthesized video data and generated match articles and reports

[0524] Operation: The server provides the synthesized video data and generated match articles and spectator reports to the user's device, as well as a means to share them on external platforms such as social media.

[0525] Output: Composite images and articles provided to the user's device, content shared on external platforms

[0526] This allows users to enjoy video content in which professional commentary, commentary, and cheering sounds are combined with footage of sports events they have filmed themselves, and they can also easily share this content with other people.

[0527] 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.

[0528] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds from video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0529] User device:

[0530] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0531] server:

[0532] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0533] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[0534] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0535] Emotion Engine:

[0536] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0537] Automatic match coverage and match reports:

[0538] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0539] Examples:

[0540] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0541] The system allows users to receive professional commentary and commentary, detailed match coverage, and enjoy a more personalized viewing experience based on their emotions.

[0542] The processing flow will be explained below.

[0543] Step 1:

[0544] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0545] Step 2:

[0546] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0547] Step 3:

[0548] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0549] Step 4:

[0550] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0551] Step 5:

[0552] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0553] Step 6:

[0554] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0555] Step 7:

[0556] The server uses an emotion engine to recognize the user's emotional state, analyzing the user's facial expressions, tone of voice, touch input, and other data acquired from the device to identify emotions.

[0557] Step 8:

[0558] The server adjusts the commentary, commentary, and cheering sounds it generates based on the user's perceived emotional state, for example, increasing the tone of the commentary if the user is perceived to be happy.

[0559] Step 9:

[0560] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0561] Step 10:

[0562] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0563] Step 11:

[0564] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0565] Step 12:

[0566] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0567] Step 13:

[0568] The server also applies the user's emotional information to match articles and game reports, generating sentences that emphasize scenes that the user found particularly exciting, for example.

[0569] Step 14:

[0570] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0571] Example 2

[0572] 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."

[0573] Conventional video analysis systems for sports events have been unable to take into account the user's emotional state when automatically generating and synthesizing commentary, narration, and cheering sounds from video data, making it difficult to provide a personalized viewing experience.In addition, there has been a lack of easy ways for users to obtain detailed match articles and spectator reports based on the results of video data analysis.

[0574] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0575] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, means for recognizing a user's emotion, means for adjusting the generated commentary and commentary based on the user's emotional state, and means for providing the synthesized video data to the user terminal. This makes it possible to provide a more personalized viewing experience based on the user's emotion. Furthermore, by adding a function for automatically generating match articles and viewing reports, users can easily obtain detailed analysis information.

[0576] A "user terminal" is a device operated by a user, and refers to digital devices such as smartphones, tablets, and personal computers.

[0577] "Video data" refers to recorded video of a sports event transmitted from a user terminal.

[0578] "Server" refers to a computer system that analyzes video data received from user terminals and performs processing, generation, and synthesis.

[0579] "Analysis" refers to the process of analyzing video data to identify specific sporting events.

[0580] A "sporting event" refers to a specific occurrence during a sporting competition (e.g., a goal or a foul).

[0581] "Identification" refers to finding specific sporting events from the analyzed data.

[0582] "Live commentary" refers to the content of a sports event being explained verbally in real time.

[0583] "Explanation" refers to supplementary information such as technical aspects and background information about sports competitions.

[0584] "Cheer sounds" refers to the sounds of cheers inserted into exciting scenes such as sporting events.

[0585] "Generated" refers to the use of AI technology and algorithms to create commentary, commentary, and cheering audio.

[0586] "Synthesis" refers to the process of integrating generated commentary, commentary, and cheering sounds into the original video data.

[0587] "Emotion recognition" refers to technology that analyzes a user's emotional state from facial expressions, tone of voice, and touch input.

[0588] "Adjusting" refers to changing the tone and content of the commentary or commentary generated based on the user's emotional state.

[0589] A "match report" refers to an article that describes important scenes or events in a sports competition in written form.

[0590] "Spectator report" refers to a report that includes detailed analytical information about a sports competition.

[0591] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds for video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0592] User terminal

[0593] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0594] server

[0595] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques and pattern recognition algorithms (e.g., OpenCV, TensorFlow), and records the event timestamps.

[0596] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to generate realistic commentary and commentary. The generated audio is then converted to a realistic form using a speech generation module (e.g., Amazon Polly, Google Text-to-Speech).

[0597] The server then uses a video editing module (e.g., FFmpeg) to properly place commentary, commentary, and cheering audio into the video, enhancing the sense of realism.

[0598] Emotion Engine

[0599] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. This recognition uses facial expression analysis technology (e.g., dlib), voice tone analysis technology, and touch input analysis technology. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0600] Automatic generation of match articles and match reports

[0601] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0602] Specific examples

[0603] For example, suppose a user films a child's soccer game and uploads the video. The server analyzes the video data and detects specific events such as goals and fouls. When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narration will become more excited.

[0604] Furthermore, based on the analysis results of the entire match, match articles such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game" are automatically generated.

[0605] Prompt Sentence Examples

[0606] Examples of prompts to be input to a generative AI model include:

[0607] "We analyzed a video of a children's soccer game. There was a goal. Please generate audio commentary such as 'Great shot!' and cheers for the goal. Also, since the user's emotions are very happy, please adjust the tone of the commentary to be more excited."

[0608] The system allows users to receive professional commentary and commentary, detailed match coverage, and a personalized viewing experience tailored to their emotions.

[0609] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0610] Step 1:

[0611] Upload video data from the user's device

[0612] Description: Users access a dedicated application or web interface using their own devices such as smartphones or PCs. After selecting the video data and pressing the upload button, the video data is sent to the server.

[0613] Input: Video data file

[0614] Output: Notification of completion of video data transmission to the server

[0615] Specific operation: The user taps the "Upload Video" button in the app, selects a video file from the device's storage, and then presses the "Upload" button to send the file to the server.

[0616] Step 2:

[0617] Video data analysis by the server

[0618] Description: The server analyzes the received video data using computer vision techniques and pattern recognition algorithms to identify specific sporting events (e.g. goals and fouls) and record them with timestamps.

[0619] Input: Uploaded video data

[0620] Output: Sport event identification results and timestamps

[0621] How it works: Once the video data arrives at the server, the video analysis process begins automatically. The server analyzes each frame, tracking player movements and ball position to detect events like goals and fouls. Detected events are then stored in a database with a timestamp.

[0622] Step 3:

[0623] Server-based generation of commentary, commentary, and cheering sounds

[0624] Description: Based on the identified event information, the server generates corresponding commentary, commentary, and cheering audio using a generative AI model (e.g., GPT-4), and uses speech synthesis technology (e.g., Amazon Polly, Google Text-to-Speech) to create an audio file from the generated text.

[0625] Input: Sport event identification results, timestamp

[0626] Output: Translated commentary and cheers text and audio files

[0627] Specific operation: For each event, the server inputs the following prompt sentence into the AI ​​model: "A goal scene has been detected. Please generate commentary text that says, 'Great shot!'" The generated text is then passed to a speech synthesis library to create an audio file.

[0628] Step 4:

[0629] Server-based audio and video data synthesis

[0630] Description: The server synthesizes the generated audio file into the video data. Using a video editing module (e.g., FFmpeg), it places the commentary, commentary, and cheering audio appropriately into the video.

[0631] Input: Original video data, generated audio file, timestamp

[0632] Output: Composite video file

[0633] How it works: Once the speech synthesis is complete, the server uses FFmpeg to add the audio file to the original video based on the timestamp. For example, a commentary saying "Great shot!" could be placed at the timestamp of a goal, creating a more immersive video.

[0634] Step 5:

[0635] Emotion recognition and feedback for users using an emotion engine

[0636] Description: The emotion engine analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state, and adjusts the tone of the commentary or commentary it generates based on the recognized emotional state.

[0637] Input: User facial expression data, tone of voice data, touch input data

[0638] Output: User's emotional state, adjusted commentary text

[0639] How it works: The emotion engine analyzes the user's previous viewing behavior and feedback. For example, if the user is very happy, the next commentary generated will be set to a higher level of excitement.

[0640] Step 6:

[0641] Automatic generation of match articles and match reports

[0642] Description: The server uses natural language generation AI to automatically generate match articles and match reports based on the analysis results and identified event information.

[0643] Input: Sporting event identification results

[0644] Output: Automatically generated match articles and match reports

[0645] What it does: The server inputs details of a sporting event into the AI ​​model and generates "match highlights and articles," such as "Player A scored a spectacular goal in the fifth minute of the first half, deciding the game."

[0646] Step 7:

[0647] Delivery of results to user devices

[0648] Description: The server delivers the generated video and match reports to the user's device, allowing the user to enjoy professional commentary, commentary, and detailed reports.

[0649] Input: Composite video files, automatically generated match reports and game reports

[0650] Output: Notification and download link to user's device

[0651] What happens: The server sends the created video file and article to the user via the app or web interface. The user logs in to download or watch the content. The app displays a notification saying "New video available," which the user can tap to go to the viewing screen.

[0652] The system provides users with a personalized and professional viewing experience.

[0653] (Application example 2)

[0654] 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."

[0655] Conventional sports viewing systems lack the means to automatically generate and synthesize realistic commentary, commentary, and cheering sounds for user-recorded video, resulting in a less than satisfactory viewing experience. It is also difficult to provide personalized content that reflects the user's emotions. This can result in a decrease in the viewer's sense of immersion and satisfaction.

[0656] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, narration, and cheering sounds for the identified sporting event, means for recognizing the user's emotion and adjusting the commentary, narration, and cheering sounds based on the emotion, means for synthesizing the generated commentary, narration, and cheering sounds with the video data, and means for providing the synthesized video data to the user terminal. This allows the user to enjoy video with professional commentary and narration, and a viewing experience personalized according to their emotion.

[0657] "User terminal" refers to digital devices such as smartphones, tablets, and personal computers operated by users.

[0658] "Video Data" refers to data in digital format that includes footage of a sports competition.

[0659] A "sporting event" refers to a specific significant moment or occurrence in a sporting competition, such as a goal or a foul.

[0660] "Live commentary" refers to audio commentary that explains and reports the situation and plays during a sports competition in real time.

[0661] "Commentary" refers to audio commentary that includes detailed explanations of the content, tactics, and player movements of a sports competition.

[0662] "Cheer sounds" refers to sounds that simulate the cheers of spectators, expressing the excitement and thrill of moments at sporting events.

[0663] "Means for recognizing emotions" refers to a system that analyzes and identifies a user's emotional state from facial images, tone of voice, touch input, etc.

[0664] "Means for adjusting commentary, commentary, and cheering sounds" refers to a system that changes the content and tone of the commentary, commentary, and cheering sounds generated based on the user's emotional state.

[0665] "Means for synthesis" refers to the technology for overlaying the generated commentary, commentary and cheering sounds onto the original video data.

[0666] This invention is a system that automatically generates and synthesizes commentary, narration, and cheering sounds for video data of sports events by operating multiple modules in coordination, providing users with a personalized viewing experience. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes an emotion engine.

[0667] User terminal

[0668] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. The user selects the video data using a dedicated application and operates the upload button, which sends the data to the server.

[0669] server

[0670] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques (e.g., OpenCV) and pattern recognition algorithms, and records the event timestamps.

[0671] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT model) is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[0672] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0673] Emotion Engine

[0674] The emotion engine is a module for recognizing the user's emotions. It analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state. Based on the emotional state recognized by the emotion engine, the server adjusts the generated commentary, narration, and cheering audio accordingly. This process uses the Emotion Recognizer library (e.g., EmotionRecognizer).

[0675] Automatic generation of match articles and match reports

[0676] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used to generate text that describes the highlights and important scenes of the match in detail. The generated match articles and spectator reports are also provided to the user's device.

[0677] Specific examples

[0678] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0679] Prompt Sentence Examples

[0680] An example of a prompt for a generative AI model is, "The user is very excited. Please commentate by saying, 'Great shot!' when a goal is scored."

[0681] This allows users to enjoy professional commentary and commentary, as well as a personalized viewing experience tailored to their emotions.

[0682] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0683] Step 1:

[0684] Uploading video data from user devices

[0685] Users use digital devices such as smartphones, tablets, and PCs to upload video data of sports events they have taken to a server via a dedicated application. The input is the video file taken by the user, and the output is the video file being transferred to the server.

[0686] Step 2:

[0687] Video data analysis

[0688] The server analyzes the received video data and identifies specific sporting events. This analysis uses computer vision techniques such as OpenCV and pattern recognition algorithms. The input is the uploaded video file, and the output is the timestamp information of the identified sporting events. Specifically, it sequentially scans each frame of the video data to detect characteristic events such as goals and fouls.

[0689] Step 3:

[0690] User Emotion Recognition

[0691] The server analyzes the user's facial expressions and tone of voice to recognize the user's emotions. The EmotionRecognizer library is used here. The input is the user's facial image and voice data, and the output is the recognized emotional state (e.g., joy, excitement). Specifically, it extracts features from the user's facial image and inputs them into a model to estimate emotions.

[0692] Step 4:

[0693] Generation of commentary, commentary and cheering sounds

[0694] The server generates commentary and cheering audio based on the identified sporting event and the user's emotional state. This process utilizes natural language generation AI (e.g., GPT model). The input is the timestamp information of the sporting event and the user's emotional state, and the output is a specific audio file. Specifically, the template sentence corresponding to the identified event is used as the prompt sentence, and the generative AI model generates natural commentary and commentary, which is then converted into audio.

[0695] Step 5:

[0696] Synthesizing live commentary, commentary, and cheering sounds into video data

[0697] The server synthesizes the generated commentary, commentary, and cheering audio into the original video data. This is done using a video editing module. The input is the original video file and the generated audio file, and the output is the synthesized video data. Specifically, the audio file is inserted and synchronized at the appropriate position in the video file according to the timestamp.

[0698] Step 6:

[0699] Providing the synthesized video data to the user's device

[0700] The server distributes the synthesized video data to the user terminal. The input is the synthesized video file, and the output is a video file encoded in a format that can be played on the user terminal. Specifically, the video data is provided to the user terminal in streaming or download format.

[0701] Step 7:

[0702] Automatic generation of match articles and match reports

[0703] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used here. The input is the timestamp information and related detailed data of the sporting event, and the output is the generated text report. Specifically, it generates prompt sentences based on the highlights of each event, and then uses the natural language generation AI to create detailed sentences.

[0704] Prompt Sentence Examples

[0705] The prompt sentence is, "The user is very excited. Please commentate when the goal is scored, saying, 'Great shot!'"

[0706] 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.

[0707] 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.

[0708] 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.

[0709] [Third embodiment]

[0710] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0711] 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.

[0712] 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).

[0713] 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.

[0714] 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.

[0715] 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).

[0716] 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.

[0717] 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.

[0718] 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.

[0719] 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.

[0720] 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.

[0721] 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."

[0722] This invention is a form for realizing a system that uses video data of sports events filmed by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0723] User device:

[0724] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0725] server:

[0726] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0727] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form via a speech generation module.

[0728] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0729] Additionally, the server automatically generates match articles and match reports based on the identified sporting event information, similarly using natural language generation AI to create detailed descriptions of the match's highlights and key moments.

[0730] Examples:

[0731] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0732] This allows users to receive professional commentary and commentary along with detailed match coverage, providing a great sports experience.

[0733] The processing flow will be explained below.

[0734] Step 1:

[0735] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0736] Step 2:

[0737] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0738] Step 3:

[0739] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0740] Step 4:

[0741] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0742] Step 5:

[0743] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0744] Step 6:

[0745] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0746] Step 7:

[0747] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0748] Step 8:

[0749] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0750] Step 9:

[0751] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0752] Step 10:

[0753] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0754] Step 11:

[0755] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0756] Example 1

[0757] 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."

[0758] In recent years, there has been an increase in opportunities to film sporting events using digital devices such as smartphones and tablets. However, the filmed videos often lack a sense of realism for viewers because they are not accompanied by professional commentary, narration, or cheering sounds. Furthermore, creating match articles or spectator reports based on filmed footage requires a great deal of time and effort. To solve this problem, a system is needed that can automatically generate commentary, narration, and cheering sounds from filmed video data and synthesize them into the video data.

[0759] 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.

[0760] In this invention, the server includes means for receiving video data of a sports event from a user terminal, means for analyzing the received video data using computer vision technology and a pattern recognition algorithm to identify a specific sports event, means for generating commentary, narration, and cheering sounds for the identified sports event using natural language generation, means for converting the generated commentary, narration, and cheering sounds into an audio file using an audio generation system and synthesizing the audio with the video data, means for providing the synthesized video data to the user terminal, means for automatically generating a match article and a spectator report using natural language generation based on information about the identified sports event, and means for providing the generated match article and spectator report to the user terminal. This makes it possible to automatically add professional commentary, narration, and cheering sounds to filmed sports video, and further generate a match article and a spectator report.

[0761] A "user terminal" is a digital device used by a user, such as a smartphone, tablet, or PC.

[0762] "Video data" is digital data containing moving images of a sporting event.

[0763] A "server" is a computer system that analyzes, processes, and stores data received from user terminals.

[0764] "Computer vision technology" is a technology that allows computers to analyze images and videos and understand their meaning.

[0765] A "pattern recognition algorithm" is a mathematical technique for identifying specific patterns or features in data.

[0766] "Sporting event" refers to a specific action or occurrence in a sporting competition (e.g., a goal, a foul).

[0767] "Natural language generation" is a technology that allows a computer to automatically create natural-sounding sentences based on data.

[0768] "Commentary" refers to audio or text that verbally describes the progress of a sporting event or important events.

[0769] "Cheering sounds" are sounds that imitate the cheers of spectators at a sporting event.

[0770] A "voice generation system" is a technology for converting text data into voice data.

[0771] An "audio file" is audio data stored in digital format.

[0772] A "video editing module" is software for integrating multiple audio and video data into a single viewable format.

[0773] A "match report" is a written report of the results and highlights of a sporting event.

[0774] A "game report" is a piece of writing in which a spectator gives a detailed description of their impressions of the game and important events that occurred.

[0775] This invention is a system that uses video data of sports events taken by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[0776] User terminal

[0777] User devices are digital devices such as smartphones, tablets, and PCs. Users use a dedicated application or web interface to upload video data of sports events they have filmed to a server. For example, a user selects a video of their child's soccer game in the dedicated application and presses the upload button, sending the data to the server.

[0778] server

[0779] The server analyzes the video data received from the user device using computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD). The server identifies specific sporting events (e.g., goals and fouls) in each video frame and records the timestamps of those events.

[0780] Once the analysis is complete, the server generates commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to create realistic commentary and commentary. For example, if a goal is detected, commentary text such as "Player A scores a great goal!" is generated. The generated text is then converted into an audio file using a speech generation system (e.g., WaveNet, Tacotron 2).

[0781] The server then uses a video editing module (e.g., FFmpeg, Adobe Premiere Pro) to combine the generated audio files with the original video data. The commentary, commentary, and cheering audio are placed appropriately based on the video timestamps, resulting in a professional-looking result. The final combined video data is then saved.

[0782] Furthermore, the server uses natural language generation AI (e.g., GPT-4) to create match articles and match reports based on the identified sporting event information, such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0783] This allows users to receive professional commentary and commentary along with detailed match coverage.

[0784] Specific examples

[0785] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (e.g., goals scored or fouls scored). When a goal is detected, a corresponding commentary such as "What a great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as, "Player A scored a great goal in the fifth minute of the first half, deciding the game."

[0786] Prompt Sentence Examples

[0787] "A user has uploaded a video of their child's soccer match. Please analyze the goal and foul scenes, generate commentary, narration, and cheering sounds based on that, and synthesize them into the video. You will also need to create a match article that details the highlights and important moments of the match."

[0788] This system allows users to easily obtain professional-quality sports footage and match reports, allowing them to enjoy a more immersive sports experience.

[0789] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0790] Step 1:

[0791] The user uploads video data to the server. Using a smartphone, tablet, or PC, the user opens a dedicated application or web interface. Next, the user selects the video data of a sports event using the file selection function and operates the upload button to send the data to the server. The input here is the video data shot by the user, and the output is the data uploaded to the server.

[0792] Step 2:

[0793] The server analyzes the received video data. The server uses computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD) to analyze the video data frame by frame. Specifically, the server divides the video data into frames, identifies specific sporting events (e.g., goal scenes or foul scenes) in each frame, and records the timestamps of those events. The input is the video data uploaded to the server, and the output is a list of specific sporting events and their timestamps.

[0794] Step 3:

[0795] The server generates commentary, narration, and cheering audio. Based on the identified sporting events, the server uses natural language generation AI (e.g., GPT-4) to generate commentary, narration, and cheering audio text. For example, if a goal is detected, the server generates commentary text such as "Player A scores a great goal!". The server then uses a speech generation system (e.g., WaveNet, Tacotron 2) to convert this text into an audio file. The input is a list of specific sporting events and their timestamps, and the output is the generated audio files of commentary, narration, and cheering audio.

[0796] Step 4:

[0797] The server synthesizes the audio into the video. Using a video editing module (e.g., FFmpeg, Adobe Premiere Pro), the server synthesizes the generated audio file with the original video data. Specifically, the server appropriately positions the generated commentary, narration, and cheering audio based on the corresponding timestamps in the video, providing a realistic sense of presence. For example, a commentary saying "Great shot!" is inserted at the right time just before a goal is scored. The input is the original video data and the generated audio file, and the output is the final synthesized video data.

[0798] Step 5:

[0799] The server generates match articles and viewing reports. Based on the information of the identified sporting event, the server uses natural language generation AI (e.g., GPT-4) to create match articles and viewing reports. For example, it generates a sentence such as, "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game." The input is information about a specific sporting event, and the output is the generated match articles and viewing reports.

[0800] Step 6:

[0801] The server provides the generated match articles and observation reports to the user terminal. The server sends the synthesized video data and the generated match articles and observation reports to the user terminal. The user can view and enjoy this data using the terminal. The input is the synthesized video data, match articles, and observation reports, and the output is the final data provided to the user terminal.

[0802] (Application example 1)

[0803] 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."

[0804] Conventional sports video simply provides filmed footage, lacking a sense of realism due to the lack of commentary, commentary, or cheering sounds. It also faces the problem of being unable to identify important scenes and events in a game and provide appropriate commentary or articles in real time. Furthermore, there is a lack of easy ways to share the generated content on social media, etc. This makes it difficult to fully bring out the appeal of sports video shot by users.

[0805] 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.

[0806] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data to identify a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, and means for sharing the generated video data, match reports, and spectator reports on external platforms such as social media. This allows users to enjoy video with professional commentary and commentary synthesized, and also makes it easy to share the generated match reports and spectator reports.

[0807] A "user terminal" is an electronic device operated by a user, and includes smartphones, tablets, personal computers, and the like.

[0808] "Video data" refers to digital data containing visual information of a sports event captured by a user.

[0809] "Analysis" refers to the process of identifying specific sporting events using computer vision technology and pattern recognition algorithms from the received video data.

[0810] A "particular sporting event" is a significant moment in the flow of a sporting event, such as a goal, a foul, or a significant play.

[0811] "Commentary" refers to the real-time verbal description and commentary of a particular sporting event.

[0812] "Commentary" means explanations that provide additional information and background about the progress or play of a sporting event.

[0813] "Cheering sounds" refers to the generation and playback of sounds that express the excitement of a game, such as the cheers and applause of spectators.

[0814] "Synthesis" refers to the process of integrating the generated commentary, commentary, and cheering sounds into the original video data to create an integrated video and audio.

[0815] A "match report" is a piece of text that provides detailed descriptions of important moments and results of a match.

[0816] A "match report" is a piece of writing that includes the overall flow and highlights of the match, as well as the spectator's impressions and opinions.

[0817] "Social media" refers to online platforms for sharing and interacting with user-generated content.

[0818] "External Platform" refers to other online services or websites that are different from this system but can be linked to it.

[0819] This invention provides a system that automatically generates commentary, commentary, and cheering sounds from video data of sports events taken by users, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module.

[0820] User terminal

[0821] The user devices are digital devices such as smartphones, tablets, and PCs, and users use them to capture video data of sports events and upload it to a server via a dedicated application or web interface. The video data captured by users is stored in cloud storage (e.g., Amazon S3).

[0822] server

[0823] The server has the function of analyzing the video data stored in the cloud storage. First, it analyzes the received video data using computer vision technology (e.g., OpenCV) to identify specific sporting events (goals, fouls, important plays, etc.). During this analysis process, timestamps of specific events are recorded.

[0824] Next, based on the identified events, a generative AI model (e.g., GPT-4) is used to generate commentary, commentary, and cheers, and the generated text is output as an audio file via a speech synthesis API (e.g., Google Cloud Text-to-Speech).

[0825] The generated commentary, commentary, and cheering audio is then combined with the original video data using the API of video editing software (e.g., Adobe Premiere Pro) to create professional video content.

[0826] The server also uses the generative AI model to generate match articles and game reports based on the analysis results. These reports are delivered to users' devices and can be shared on external platforms such as social media.

[0827] Specific examples

[0828] For example, a user films a child's soccer game and uploads the footage to a server via a smartphone application. The server analyzes the video data and identifies specific events (e.g., a goal). For the identified event, a generative AI model generates commentary text such as "Great shot!" and a voice synthesis API is used to create an audio file. The generated audio is then combined with the original video, creating immersive video content. The generative AI model is also used to create match articles and spectator reports, providing text such as "Player A scored a brilliant goal in the fifth minute of the first half, sealing the victory."

[0829] Prompt Sentence Examples

[0830] 1. Video analysis prompt:

[0831] "Please identify the important sporting events (goals, fouls, etc.) in this video."

[0832] 2. Commentary generation prompts:

[0833] "Generate commentary text for goals scored in a match. For example, 'Great shot!'"

[0834] 3. Match article generation prompt:

[0835] "Generate an article detailing the highlights and key moments of the match. For example, 'Player A scored a brilliant goal in the fifth minute of the first half, deciding the game.'"

[0836] As described above, this system enables users to easily create and share professional sports video content.

[0837] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0838] Step 1:

[0839] Input: Video data of sports events taken by the user

[0840] How it works: Users use their smartphone, tablet, or PC to upload footage of sports events to a server via a dedicated application or web interface.

[0841] Output: Video data stored in cloud storage (e.g. Amazon S3)

[0842] Step 2:

[0843] Input: Video data stored in cloud storage

[0844] How it works: The server retrieves video data stored in cloud storage and analyzes the video using computer vision techniques (e.g., OpenCV). This analysis identifies specific sporting events (e.g., goals, fouls, key plays) and captures event timestamps.

[0845] Output: timestamp and information about identified sporting events

[0846] Step 3:

[0847] Input: Timestamp and information of identified sporting events

[0848] How it works: The server uses a generative AI model (e.g., GPT-4) to generate commentary, commentary, and cheers for the identified events. Depending on the text you want to generate, input the following prompts into the generative AI model:

[0849] Example prompt: "Generate commentary text for a goal in a game. For example, 'Great shot!'"

[0850] Output: Text of the generated commentary, commentary, and cheers

[0851] Step 4:

[0852] Input: Generated commentary, commentary, and cheering audio text

[0853] How it works: The server uses a speech synthesis API (e.g., Google Cloud Text-to-Speech) to output the generated commentary, commentary, and cheering text as an audio file.

[0854] Output: Generated audio file

[0855] Step 5:

[0856] Input: Video data and generated audio files

[0857] How it works: The server uses the API of video editing software (e.g. Adobe Premiere Pro) to synthesize the generated commentary, commentary, and cheering audio into the original video data.

[0858] Output: Composite video data

[0859] Step 6:

[0860] Input: Timestamp and information of identified sporting events

[0861] How it works: The server uses the generative AI model to generate match articles and game reports. Based on the analysis results, it inputs the following prompt sentences into the generative AI model:

[0862] Example prompt: "Generate an article detailing the highlights and key moments of a match. For example, 'Player A scored a spectacular goal in the fifth minute of the first half, deciding the game.'"

[0863] Output: Generated match articles and match reports

[0864] Step 7:

[0865] Input: Synthesized video data and generated match articles and reports

[0866] Operation: The server provides the synthesized video data and generated match articles and spectator reports to the user's device, as well as a means to share them on external platforms such as social media.

[0867] Output: Composite images and articles provided to the user's device, content shared on external platforms

[0868] This allows users to enjoy video content in which professional commentary, commentary, and cheering sounds are combined with footage of sports events they have filmed themselves, and they can also easily share this content with other people.

[0869] 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.

[0870] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds from video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0871] User device:

[0872] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0873] server:

[0874] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[0875] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[0876] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[0877] Emotion Engine:

[0878] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0879] Automatic match coverage and match reports:

[0880] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0881] Examples:

[0882] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[0883] The system allows users to receive professional commentary and commentary, detailed match coverage, and enjoy a more personalized viewing experience based on their emotions.

[0884] The processing flow will be explained below.

[0885] Step 1:

[0886] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[0887] Step 2:

[0888] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[0889] Step 3:

[0890] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[0891] Step 4:

[0892] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[0893] Step 5:

[0894] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[0895] Step 6:

[0896] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[0897] Step 7:

[0898] The server uses an emotion engine to recognize the user's emotional state, analyzing the user's facial expressions, tone of voice, touch input, and other data acquired from the device to identify emotions.

[0899] Step 8:

[0900] The server adjusts the commentary, commentary, and cheering sounds it generates based on the user's perceived emotional state, for example, increasing the tone of the commentary if the user is perceived to be happy.

[0901] Step 9:

[0902] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[0903] Step 10:

[0904] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[0905] Step 11:

[0906] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[0907] Step 12:

[0908] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[0909] Step 13:

[0910] The server also applies the user's emotional information to match articles and game reports, generating sentences that emphasize scenes that the user found particularly exciting, for example.

[0911] Step 14:

[0912] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[0913] Example 2

[0914] 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."

[0915] Conventional video analysis systems for sports events have been unable to take into account the user's emotional state when automatically generating and synthesizing commentary, narration, and cheering sounds from video data, making it difficult to provide a personalized viewing experience.In addition, there has been a lack of easy ways for users to obtain detailed match articles and spectator reports based on the results of video data analysis.

[0916] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0917] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, means for recognizing a user's emotion, means for adjusting the generated commentary and commentary based on the user's emotional state, and means for providing the synthesized video data to the user terminal. This makes it possible to provide a more personalized viewing experience based on the user's emotion. Furthermore, by adding a function for automatically generating match articles and viewing reports, users can easily obtain detailed analysis information.

[0918] A "user terminal" is a device operated by a user, and refers to digital devices such as smartphones, tablets, and personal computers.

[0919] "Video data" refers to recorded video of a sports event transmitted from a user terminal.

[0920] "Server" refers to a computer system that analyzes video data received from user terminals and performs processing, generation, and synthesis.

[0921] "Analysis" refers to the process of analyzing video data to identify specific sporting events.

[0922] A "sporting event" refers to a specific occurrence during a sporting competition (e.g., a goal or a foul).

[0923] "Identification" refers to finding specific sporting events from the analyzed data.

[0924] "Live commentary" refers to the content of a sports event being explained verbally in real time.

[0925] "Explanation" refers to supplementary information such as technical aspects and background information about sports competitions.

[0926] "Cheer sounds" refers to the sounds of cheers inserted into exciting scenes such as sporting events.

[0927] "Generated" refers to the use of AI technology and algorithms to create commentary, commentary, and cheering audio.

[0928] "Synthesis" refers to the process of integrating generated commentary, commentary, and cheering sounds into the original video data.

[0929] "Emotion recognition" refers to technology that analyzes a user's emotional state from facial expressions, tone of voice, and touch input.

[0930] "Adjusting" refers to changing the tone and content of the commentary or commentary generated based on the user's emotional state.

[0931] A "match report" refers to an article that describes important scenes or events in a sports competition in written form.

[0932] "Spectator report" refers to a report that includes detailed analytical information about a sports competition.

[0933] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds for video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[0934] User terminal

[0935] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[0936] server

[0937] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques and pattern recognition algorithms (e.g., OpenCV, TensorFlow), and records the event timestamps.

[0938] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to generate realistic commentary and commentary. The generated audio is then converted to a realistic form using a speech generation module (e.g., Amazon Polly, Google Text-to-Speech).

[0939] The server then uses a video editing module (e.g., FFmpeg) to properly place commentary, commentary, and cheering audio into the video, enhancing the sense of realism.

[0940] Emotion Engine

[0941] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. This recognition uses facial expression analysis technology (e.g., dlib), voice tone analysis technology, and touch input analysis technology. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[0942] Automatic generation of match articles and match reports

[0943] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[0944] Specific examples

[0945] For example, suppose a user films a child's soccer game and uploads the video. The server analyzes the video data and detects specific events such as goals and fouls. When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narration will become more excited.

[0946] Furthermore, based on the analysis results of the entire match, match articles such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game" are automatically generated.

[0947] Prompt Sentence Examples

[0948] Examples of prompts to be input to a generative AI model include:

[0949] "We analyzed a video of a children's soccer game. There was a goal. Please generate audio commentary such as 'Great shot!' and cheers for the goal. Also, since the user's emotions are very happy, please adjust the tone of the commentary to be more excited."

[0950] The system allows users to receive professional commentary and commentary, detailed match coverage, and a personalized viewing experience tailored to their emotions.

[0951] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0952] Step 1:

[0953] Upload video data from the user's device

[0954] Description: Users access a dedicated application or web interface using their own devices such as smartphones or PCs. After selecting the video data and pressing the upload button, the video data is sent to the server.

[0955] Input: Video data file

[0956] Output: Notification of completion of video data transmission to the server

[0957] Specific operation: The user taps the "Upload Video" button in the app, selects a video file from the device's storage, and then presses the "Upload" button to send the file to the server.

[0958] Step 2:

[0959] Video data analysis by the server

[0960] Description: The server analyzes the received video data using computer vision techniques and pattern recognition algorithms to identify specific sporting events (e.g. goals and fouls) and record them with timestamps.

[0961] Input: Uploaded video data

[0962] Output: Sport event identification results and timestamps

[0963] How it works: Once the video data arrives at the server, the video analysis process begins automatically. The server analyzes each frame, tracking player movements and ball position to detect events like goals and fouls. Detected events are then stored in a database with a timestamp.

[0964] Step 3:

[0965] Server-based generation of commentary, commentary, and cheering sounds

[0966] Description: Based on the identified event information, the server generates corresponding commentary, commentary, and cheering audio using a generative AI model (e.g., GPT-4), and uses speech synthesis technology (e.g., Amazon Polly, Google Text-to-Speech) to create an audio file from the generated text.

[0967] Input: Sport event identification results, timestamp

[0968] Output: Translated commentary and cheers text and audio files

[0969] Specific operation: For each event, the server inputs the following prompt sentence into the AI ​​model: "A goal scene has been detected. Please generate commentary text that says, 'Great shot!'" The generated text is then passed to a speech synthesis library to create an audio file.

[0970] Step 4:

[0971] Server-based audio and video data synthesis

[0972] Description: The server synthesizes the generated audio file into the video data. Using a video editing module (e.g., FFmpeg), it places the commentary, commentary, and cheering audio appropriately into the video.

[0973] Input: Original video data, generated audio file, timestamp

[0974] Output: Composite video file

[0975] How it works: Once the speech synthesis is complete, the server uses FFmpeg to add the audio file to the original video based on the timestamp. For example, a commentary saying "Great shot!" could be placed at the timestamp of a goal, creating a more immersive video.

[0976] Step 5:

[0977] Emotion recognition and feedback for users using an emotion engine

[0978] Description: The emotion engine analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state, and adjusts the tone of the commentary or commentary it generates based on the recognized emotional state.

[0979] Input: User facial expression data, tone of voice data, touch input data

[0980] Output: User's emotional state, adjusted commentary text

[0981] How it works: The emotion engine analyzes the user's previous viewing behavior and feedback. For example, if the user is very happy, the next commentary generated will be set to a higher level of excitement.

[0982] Step 6:

[0983] Automatic generation of match articles and match reports

[0984] Description: The server uses natural language generation AI to automatically generate match articles and match reports based on the analysis results and identified event information.

[0985] Input: Sporting event identification results

[0986] Output: Automatically generated match articles and match reports

[0987] What it does: The server inputs details of a sporting event into the AI ​​model and generates "match highlights and articles," such as "Player A scored a spectacular goal in the fifth minute of the first half, deciding the game."

[0988] Step 7:

[0989] Delivery of results to user devices

[0990] Description: The server delivers the generated video and match reports to the user's device, allowing the user to enjoy professional commentary, commentary, and detailed reports.

[0991] Input: Composite video files, automatically generated match reports and game reports

[0992] Output: Notification and download link to user's device

[0993] What happens: The server sends the created video file and article to the user via the app or web interface. The user logs in to download or watch the content. The app displays a notification saying "New video available," which the user can tap to go to the viewing screen.

[0994] The system provides users with a personalized and professional viewing experience.

[0995] (Application example 2)

[0996] 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."

[0997] Conventional sports viewing systems lack the means to automatically generate and synthesize realistic commentary, commentary, and cheering sounds for user-recorded video, resulting in a less than satisfactory viewing experience. It is also difficult to provide personalized content that reflects the user's emotions. This can result in a decrease in the viewer's sense of immersion and satisfaction.

[0998] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, narration, and cheering sounds for the identified sporting event, means for recognizing the user's emotion and adjusting the commentary, narration, and cheering sounds based on the emotion, means for synthesizing the generated commentary, narration, and cheering sounds with the video data, and means for providing the synthesized video data to the user terminal. This allows the user to enjoy video with professional commentary and narration, and a viewing experience personalized according to their emotion.

[0999] "User terminal" refers to digital devices such as smartphones, tablets, and personal computers operated by users.

[1000] "Video Data" refers to data in digital format that includes footage of a sports competition.

[1001] A "sporting event" refers to a specific significant moment or occurrence in a sporting competition, such as a goal or a foul.

[1002] "Live commentary" refers to audio commentary that explains and reports the situation and plays during a sports competition in real time.

[1003] "Commentary" refers to audio commentary that includes detailed explanations of the content, tactics, and player movements of a sports competition.

[1004] "Cheer sounds" refers to sounds that simulate the cheers of spectators, expressing the excitement and thrill of moments at sporting events.

[1005] "Means for recognizing emotions" refers to a system that analyzes and identifies a user's emotional state from facial images, tone of voice, touch input, etc.

[1006] "Means for adjusting commentary, commentary, and cheering sounds" refers to a system that changes the content and tone of the commentary, commentary, and cheering sounds generated based on the user's emotional state.

[1007] "Means for synthesis" refers to the technology for overlaying the generated commentary, commentary and cheering sounds onto the original video data.

[1008] This invention is a system that automatically generates and synthesizes commentary, narration, and cheering sounds for video data of sports events by operating multiple modules in coordination, providing users with a personalized viewing experience. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes an emotion engine.

[1009] User terminal

[1010] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. The user selects the video data using a dedicated application and operates the upload button, which sends the data to the server.

[1011] server

[1012] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques (e.g., OpenCV) and pattern recognition algorithms, and records the event timestamps.

[1013] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT model) is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[1014] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[1015] Emotion Engine

[1016] The emotion engine is a module for recognizing the user's emotions. It analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state. Based on the emotional state recognized by the emotion engine, the server adjusts the generated commentary, narration, and cheering audio accordingly. This process uses the Emotion Recognizer library (e.g., EmotionRecognizer).

[1017] Automatic generation of match articles and match reports

[1018] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used to generate text that describes the highlights and important scenes of the match in detail. The generated match articles and spectator reports are also provided to the user's device.

[1019] Specific examples

[1020] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[1021] Prompt Sentence Examples

[1022] An example of a prompt for a generative AI model is, "The user is very excited. Please commentate by saying, 'Great shot!' when a goal is scored."

[1023] This allows users to enjoy professional commentary and commentary, as well as a personalized viewing experience tailored to their emotions.

[1024] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1025] Step 1:

[1026] Uploading video data from user devices

[1027] Users use digital devices such as smartphones, tablets, and PCs to upload video data of sports events they have taken to a server via a dedicated application. The input is the video file taken by the user, and the output is the video file being transferred to the server.

[1028] Step 2:

[1029] Video data analysis

[1030] The server analyzes the received video data and identifies specific sporting events. This analysis uses computer vision techniques such as OpenCV and pattern recognition algorithms. The input is the uploaded video file, and the output is the timestamp information of the identified sporting events. Specifically, it sequentially scans each frame of the video data to detect characteristic events such as goals and fouls.

[1031] Step 3:

[1032] User Emotion Recognition

[1033] The server analyzes the user's facial expressions and tone of voice to recognize the user's emotions. The EmotionRecognizer library is used here. The input is the user's facial image and voice data, and the output is the recognized emotional state (e.g., joy, excitement). Specifically, it extracts features from the user's facial image and inputs them into a model to estimate emotions.

[1034] Step 4:

[1035] Generation of commentary, commentary and cheering sounds

[1036] The server generates commentary and cheering audio based on the identified sporting event and the user's emotional state. This process utilizes natural language generation AI (e.g., GPT model). The input is the timestamp information of the sporting event and the user's emotional state, and the output is a specific audio file. Specifically, the template sentence corresponding to the identified event is used as the prompt sentence, and the generative AI model generates natural commentary and commentary, which is then converted into audio.

[1037] Step 5:

[1038] Synthesizing live commentary, commentary, and cheering sounds into video data

[1039] The server synthesizes the generated commentary, commentary, and cheering audio into the original video data. This is done using a video editing module. The input is the original video file and the generated audio file, and the output is the synthesized video data. Specifically, the audio file is inserted and synchronized at the appropriate position in the video file according to the timestamp.

[1040] Step 6:

[1041] Providing the synthesized video data to the user's device

[1042] The server distributes the synthesized video data to the user terminal. The input is the synthesized video file, and the output is a video file encoded in a format that can be played on the user terminal. Specifically, the video data is provided to the user terminal in streaming or download format.

[1043] Step 7:

[1044] Automatic generation of match articles and match reports

[1045] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used here. The input is the timestamp information and related detailed data of the sporting event, and the output is the generated text report. Specifically, it generates prompt sentences based on the highlights of each event, and then uses the natural language generation AI to create detailed sentences.

[1046] Prompt Sentence Examples

[1047] The prompt sentence is, "The user is very excited. Please commentate when the goal is scored, saying, 'Great shot!'"

[1048] 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.

[1049] 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.

[1050] 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.

[1051] [Fourth embodiment]

[1052] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1053] 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.

[1054] 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).

[1055] 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.

[1056] 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.

[1057] 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).

[1058] 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.

[1059] 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.

[1060] 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.

[1061] 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.

[1062] 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.

[1063] 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.

[1064] 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."

[1065] This invention is a form for realizing a system that uses video data of sports events filmed by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[1066] User device:

[1067] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[1068] server:

[1069] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[1070] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form via a speech generation module.

[1071] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[1072] Additionally, the server automatically generates match articles and match reports based on the identified sporting event information, similarly using natural language generation AI to create detailed descriptions of the match's highlights and key moments.

[1073] Examples:

[1074] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[1075] This allows users to receive professional commentary and commentary along with detailed match coverage, providing a great sports experience.

[1076] The processing flow will be explained below.

[1077] Step 1:

[1078] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[1079] Step 2:

[1080] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[1081] Step 3:

[1082] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[1083] Step 4:

[1084] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[1085] Step 5:

[1086] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[1087] Step 6:

[1088] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[1089] Step 7:

[1090] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[1091] Step 8:

[1092] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[1093] Step 9:

[1094] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[1095] Step 10:

[1096] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[1097] Step 11:

[1098] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[1099] Example 1

[1100] 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."

[1101] In recent years, there has been an increase in opportunities to film sporting events using digital devices such as smartphones and tablets. However, the filmed videos often lack a sense of realism for viewers because they are not accompanied by professional commentary, narration, or cheering sounds. Furthermore, creating match articles or spectator reports based on filmed footage requires a great deal of time and effort. To solve this problem, a system is needed that can automatically generate commentary, narration, and cheering sounds from filmed video data and synthesize them into the video data.

[1102] 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.

[1103] In this invention, the server includes means for receiving video data of a sports event from a user terminal, means for analyzing the received video data using computer vision technology and a pattern recognition algorithm to identify a specific sports event, means for generating commentary, narration, and cheering sounds for the identified sports event using natural language generation, means for converting the generated commentary, narration, and cheering sounds into an audio file using an audio generation system and synthesizing the audio with the video data, means for providing the synthesized video data to the user terminal, means for automatically generating a match article and a spectator report using natural language generation based on information about the identified sports event, and means for providing the generated match article and spectator report to the user terminal. This makes it possible to automatically add professional commentary, narration, and cheering sounds to filmed sports video, and further generate a match article and a spectator report.

[1104] A "user terminal" is a digital device used by a user, such as a smartphone, tablet, or PC.

[1105] "Video data" is digital data containing moving images of a sporting event.

[1106] A "server" is a computer system that analyzes, processes, and stores data received from user terminals.

[1107] "Computer vision technology" is a technology that allows computers to analyze images and videos and understand their meaning.

[1108] A "pattern recognition algorithm" is a mathematical technique for identifying specific patterns or features in data.

[1109] "Sporting event" refers to a specific action or occurrence in a sporting competition (e.g., a goal, a foul).

[1110] "Natural language generation" is a technology that allows a computer to automatically create natural-sounding sentences based on data.

[1111] "Commentary" refers to audio or text that verbally describes the progress of a sporting event or important events.

[1112] "Cheering sounds" are sounds that imitate the cheers of spectators at a sporting event.

[1113] A "voice generation system" is a technology for converting text data into voice data.

[1114] An "audio file" is audio data stored in digital format.

[1115] A "video editing module" is software for integrating multiple audio and video data into a single viewable format.

[1116] A "match report" is a written report of the results and highlights of a sporting event.

[1117] A "game report" is a piece of writing in which a spectator gives a detailed description of their impressions of the game and important events that occurred.

[1118] This invention is a system that uses video data of sports events taken by users to automatically generate and synthesize accompanying commentary, commentary, and cheering sounds, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that works in conjunction with the server.

[1119] User terminal

[1120] User devices are digital devices such as smartphones, tablets, and PCs. Users use a dedicated application or web interface to upload video data of sports events they have filmed to a server. For example, a user selects a video of their child's soccer game in the dedicated application and presses the upload button, sending the data to the server.

[1121] server

[1122] The server analyzes the video data received from the user device using computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD). The server identifies specific sporting events (e.g., goals and fouls) in each video frame and records the timestamps of those events.

[1123] Once the analysis is complete, the server generates commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to create realistic commentary and commentary. For example, if a goal is detected, commentary text such as "Player A scores a great goal!" is generated. The generated text is then converted into an audio file using a speech generation system (e.g., WaveNet, Tacotron 2).

[1124] The server then uses a video editing module (e.g., FFmpeg, Adobe Premiere Pro) to combine the generated audio files with the original video data. The commentary, commentary, and cheering audio are placed appropriately based on the video timestamps, resulting in a professional-looking result. The final combined video data is then saved.

[1125] Furthermore, the server uses natural language generation AI (e.g., GPT-4) to create match articles and match reports based on the identified sporting event information, such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[1126] This allows users to receive professional commentary and commentary along with detailed match coverage.

[1127] Specific examples

[1128] For example, when a user uploads a video of a child's soccer game, the server analyzes the video and detects specific events (e.g., goals scored or fouls scored). When a goal is detected, a corresponding commentary such as "What a great shot!" and cheers are generated and superimposed onto the original video. Based on the analysis of the entire game, a "match article" is automatically generated, containing content such as, "Player A scored a great goal in the fifth minute of the first half, deciding the game."

[1129] Prompt Sentence Examples

[1130] "A user has uploaded a video of their child's soccer match. Please analyze the goal and foul scenes, generate commentary, narration, and cheering sounds based on that, and synthesize them into the video. You will also need to create a match article that details the highlights and important moments of the match."

[1131] This system allows users to easily obtain professional-quality sports footage and match reports, allowing them to enjoy a more immersive sports experience.

[1132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1133] Step 1:

[1134] The user uploads video data to the server. Using a smartphone, tablet, or PC, the user opens a dedicated application or web interface. Next, the user selects the video data of a sports event using the file selection function and operates the upload button to send the data to the server. The input here is the video data shot by the user, and the output is the data uploaded to the server.

[1135] Step 2:

[1136] The server analyzes the received video data. The server uses computer vision techniques (e.g., OpenCV, TensorFlow) and pattern recognition algorithms (e.g., YOLO, SSD) to analyze the video data frame by frame. Specifically, the server divides the video data into frames, identifies specific sporting events (e.g., goal scenes or foul scenes) in each frame, and records the timestamps of those events. The input is the video data uploaded to the server, and the output is a list of specific sporting events and their timestamps.

[1137] Step 3:

[1138] The server generates commentary, narration, and cheering audio. Based on the identified sporting events, the server uses natural language generation AI (e.g., GPT-4) to generate commentary, narration, and cheering audio text. For example, if a goal is detected, the server generates commentary text such as "Player A scores a great goal!". The server then uses a speech generation system (e.g., WaveNet, Tacotron 2) to convert this text into an audio file. The input is a list of specific sporting events and their timestamps, and the output is the generated audio files of commentary, narration, and cheering audio.

[1139] Step 4:

[1140] The server synthesizes the audio into the video. Using a video editing module (e.g., FFmpeg, Adobe Premiere Pro), the server synthesizes the generated audio file with the original video data. Specifically, the server appropriately positions the generated commentary, narration, and cheering audio based on the corresponding timestamps in the video, providing a realistic sense of presence. For example, a commentary saying "Great shot!" is inserted at the right time just before a goal is scored. The input is the original video data and the generated audio file, and the output is the final synthesized video data.

[1141] Step 5:

[1142] The server generates match articles and viewing reports. Based on the information of the identified sporting event, the server uses natural language generation AI (e.g., GPT-4) to create match articles and viewing reports. For example, it generates a sentence such as, "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game." The input is information about a specific sporting event, and the output is the generated match articles and viewing reports.

[1143] Step 6:

[1144] The server provides the generated match articles and observation reports to the user terminal. The server sends the synthesized video data and the generated match articles and observation reports to the user terminal. The user can view and enjoy this data using the terminal. The input is the synthesized video data, match articles, and observation reports, and the output is the final data provided to the user terminal.

[1145] (Application example 1)

[1146] 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."

[1147] Conventional sports video simply provides filmed footage, lacking a sense of realism due to the lack of commentary, commentary, or cheering sounds. It also faces the problem of being unable to identify important scenes and events in a game and provide appropriate commentary or articles in real time. Furthermore, there is a lack of easy ways to share the generated content on social media, etc. This makes it difficult to fully bring out the appeal of sports video shot by users.

[1148] 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.

[1149] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data to identify a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, and means for sharing the generated video data, match reports, and spectator reports on external platforms such as social media. This allows users to enjoy video with professional commentary and commentary synthesized, and also makes it easy to share the generated match reports and spectator reports.

[1150] A "user terminal" is an electronic device operated by a user, and includes smartphones, tablets, personal computers, and the like.

[1151] "Video data" refers to digital data containing visual information of a sports event captured by a user.

[1152] "Analysis" refers to the process of identifying specific sporting events using computer vision technology and pattern recognition algorithms from the received video data.

[1153] A "particular sporting event" is a significant moment in the flow of a sporting event, such as a goal, a foul, or a significant play.

[1154] "Commentary" refers to the real-time verbal description and commentary of a particular sporting event.

[1155] "Commentary" means explanations that provide additional information and background about the progress or play of a sporting event.

[1156] "Cheering sounds" refers to the generation and playback of sounds that express the excitement of a game, such as the cheers and applause of spectators.

[1157] "Synthesis" refers to the process of integrating the generated commentary, commentary, and cheering sounds into the original video data to create an integrated video and audio.

[1158] A "match report" is a piece of text that provides detailed descriptions of important moments and results of a match.

[1159] A "match report" is a piece of writing that includes the overall flow and highlights of the match, as well as the spectator's impressions and opinions.

[1160] "Social media" refers to online platforms for sharing and interacting with user-generated content.

[1161] "External Platform" refers to other online services or websites that are different from this system but can be linked to it.

[1162] This invention provides a system that automatically generates commentary, commentary, and cheering sounds from video data of sports events taken by users, and also creates match articles and spectator reports. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module.

[1163] User terminal

[1164] The user devices are digital devices such as smartphones, tablets, and PCs, and users use them to capture video data of sports events and upload it to a server via a dedicated application or web interface. The video data captured by users is stored in cloud storage (e.g., Amazon S3).

[1165] server

[1166] The server has the function of analyzing the video data stored in the cloud storage. First, it analyzes the received video data using computer vision technology (e.g., OpenCV) to identify specific sporting events (goals, fouls, important plays, etc.). During this analysis process, timestamps of specific events are recorded.

[1167] Next, based on the identified events, a generative AI model (e.g., GPT-4) is used to generate commentary, commentary, and cheers, and the generated text is output as an audio file via a speech synthesis API (e.g., Google Cloud Text-to-Speech).

[1168] The generated commentary, commentary, and cheering audio is then combined with the original video data using the API of video editing software (e.g., Adobe Premiere Pro) to create professional video content.

[1169] The server also uses the generative AI model to generate match articles and game reports based on the analysis results. These reports are delivered to users' devices and can be shared on external platforms such as social media.

[1170] Specific examples

[1171] For example, a user films a child's soccer game and uploads the footage to a server via a smartphone application. The server analyzes the video data and identifies specific events (e.g., a goal). For the identified event, a generative AI model generates commentary text such as "Great shot!" and a voice synthesis API is used to create an audio file. The generated audio is then combined with the original video, creating immersive video content. The generative AI model is also used to create match articles and spectator reports, providing text such as "Player A scored a brilliant goal in the fifth minute of the first half, sealing the victory."

[1172] Prompt Sentence Examples

[1173] 1. Video analysis prompt:

[1174] "Please identify the important sporting events (goals, fouls, etc.) in this video."

[1175] 2. Commentary generation prompts:

[1176] "Generate commentary text for goals scored in a match. For example, 'Great shot!'"

[1177] 3. Match article generation prompt:

[1178] "Generate an article detailing the highlights and key moments of the match. For example, 'Player A scored a brilliant goal in the fifth minute of the first half, deciding the game.'"

[1179] As described above, this system enables users to easily create and share professional sports video content.

[1180] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1181] Step 1:

[1182] Input: Video data of sports events taken by the user

[1183] How it works: Users use their smartphone, tablet, or PC to upload footage of sports events to a server via a dedicated application or web interface.

[1184] Output: Video data stored in cloud storage (e.g. Amazon S3)

[1185] Step 2:

[1186] Input: Video data stored in cloud storage

[1187] How it works: The server retrieves video data stored in cloud storage and analyzes the video using computer vision techniques (e.g., OpenCV). This analysis identifies specific sporting events (e.g., goals, fouls, key plays) and captures event timestamps.

[1188] Output: timestamp and information about identified sporting events

[1189] Step 3:

[1190] Input: Timestamp and information of identified sporting events

[1191] How it works: The server uses a generative AI model (e.g., GPT-4) to generate commentary, commentary, and cheers for the identified events. Depending on the text you want to generate, input the following prompts into the generative AI model:

[1192] Example prompt: "Generate commentary text for a goal in a game. For example, 'Great shot!'"

[1193] Output: Text of the generated commentary, commentary, and cheers

[1194] Step 4:

[1195] Input: Generated commentary, commentary, and cheering audio text

[1196] How it works: The server uses a speech synthesis API (e.g., Google Cloud Text-to-Speech) to output the generated commentary, commentary, and cheering text as an audio file.

[1197] Output: Generated audio file

[1198] Step 5:

[1199] Input: Video data and generated audio files

[1200] How it works: The server uses the API of video editing software (e.g. Adobe Premiere Pro) to synthesize the generated commentary, commentary, and cheering audio into the original video data.

[1201] Output: Composite video data

[1202] Step 6:

[1203] Input: Timestamp and information of identified sporting events

[1204] How it works: The server uses the generative AI model to generate match articles and game reports. Based on the analysis results, it inputs the following prompt sentences into the generative AI model:

[1205] Example prompt: "Generate an article detailing the highlights and key moments of a match. For example, 'Player A scored a spectacular goal in the fifth minute of the first half, deciding the game.'"

[1206] Output: Generated match articles and match reports

[1207] Step 7:

[1208] Input: Synthesized video data and generated match articles and reports

[1209] Operation: The server provides the synthesized video data and generated match articles and spectator reports to the user's device, as well as a means to share them on external platforms such as social media.

[1210] Output: Composite images and articles provided to the user's device, content shared on external platforms

[1211] This allows users to enjoy video content in which professional commentary, commentary, and cheering sounds are combined with footage of sports events they have filmed themselves, and they can also easily share this content with other people.

[1212] 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.

[1213] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds from video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[1214] User device:

[1215] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[1216] server:

[1217] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision technology and pattern recognition algorithms, and records timestamps of the events.

[1218] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[1219] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[1220] Emotion Engine:

[1221] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[1222] Automatic match coverage and match reports:

[1223] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[1224] Examples:

[1225] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[1226] The system allows users to receive professional commentary and commentary, detailed match coverage, and enjoy a more personalized viewing experience based on their emotions.

[1227] The processing flow will be explained below.

[1228] Step 1:

[1229] The user selects video data of a sports event from the device. The user operates a dedicated application or web interface on the device to select the video file they have taken.

[1230] Step 2:

[1231] The user uploads the selected video data from the device to the server. The user clicks the upload button and the video data is sent to the server.

[1232] Step 3:

[1233] The server receives the video data from the user, starts the video analysis module, and prepares the data.

[1234] Step 4:

[1235] The server analyzes the video data to identify specific sporting events, using computer vision technology and pattern recognition algorithms to detect goals and key plays.

[1236] Step 5:

[1237] The server records the timestamp of the identified sporting event, for example, "Goal at 05:20" or "Foul at 10:15".

[1238] Step 6:

[1239] The server generates play-by-play, commentary, and cheering audio for the identified events. The server uses natural language generation AI to generate realistic play-by-play and commentary for appropriate events.

[1240] Step 7:

[1241] The server uses an emotion engine to recognize the user's emotional state, analyzing the user's facial expressions, tone of voice, touch input, and other data acquired from the device to identify emotions.

[1242] Step 8:

[1243] The server adjusts the commentary, commentary, and cheering sounds it generates based on the user's perceived emotional state, for example, increasing the tone of the commentary if the user is perceived to be happy.

[1244] Step 9:

[1245] The server generates the commentary, commentary, and cheering sounds as realistic audio files. The server uses a speech generation module to convert text to speech.

[1246] Step 10:

[1247] The server synthesizes the generated audio file with the original video data. The server uses a video editing module to synthesize the generated audio with the corresponding timestamp of the video data.

[1248] Step 11:

[1249] The server then provides the completed video data, complete with commentary, commentary, and cheering audio, to the user's device, where the user can download and view the combined video.

[1250] Step 12:

[1251] The server automatically generates match articles and spectator reports based on the identified sporting event information, and uses natural language generation AI to create detailed match articles and reports.

[1252] Step 13:

[1253] The server also applies the user's emotional information to match articles and game reports, generating sentences that emphasize scenes that the user found particularly exciting, for example.

[1254] Step 14:

[1255] The server provides the generated match articles and match reports to users' devices, where users can download, view, and share these articles and reports.

[1256] Example 2

[1257] 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."

[1258] Conventional video analysis systems for sports events have been unable to take into account the user's emotional state when automatically generating and synthesizing commentary, narration, and cheering sounds from video data, making it difficult to provide a personalized viewing experience.In addition, there has been a lack of easy ways for users to obtain detailed match articles and spectator reports based on the results of video data analysis.

[1259] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1260] In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, commentary, and cheering sounds for the identified sporting event, means for synthesizing the generated commentary, commentary, and cheering sounds with the video data, means for recognizing a user's emotion, means for adjusting the generated commentary and commentary based on the user's emotional state, and means for providing the synthesized video data to the user terminal. This makes it possible to provide a more personalized viewing experience based on the user's emotion. Furthermore, by adding a function for automatically generating match articles and viewing reports, users can easily obtain detailed analysis information.

[1261] A "user terminal" is a device operated by a user, and refers to digital devices such as smartphones, tablets, and personal computers.

[1262] "Video data" refers to recorded video of a sports event transmitted from a user terminal.

[1263] "Server" refers to a computer system that analyzes video data received from user terminals and performs processing, generation, and synthesis.

[1264] "Analysis" refers to the process of analyzing video data to identify specific sporting events.

[1265] A "sporting event" refers to a specific occurrence during a sporting competition (e.g., a goal or a foul).

[1266] "Identification" refers to finding specific sporting events from the analyzed data.

[1267] "Live commentary" refers to the content of a sports event being explained verbally in real time.

[1268] "Explanation" refers to supplementary information such as technical aspects and background information about sports competitions.

[1269] "Cheer sounds" refers to the sounds of cheers inserted into exciting scenes such as sporting events.

[1270] "Generated" refers to the use of AI technology and algorithms to create commentary, commentary, and cheering audio.

[1271] "Synthesis" refers to the process of integrating generated commentary, commentary, and cheering sounds into the original video data.

[1272] "Emotion recognition" refers to technology that analyzes a user's emotional state from facial expressions, tone of voice, and touch input.

[1273] "Adjusting" refers to changing the tone and content of the commentary or commentary generated based on the user's emotional state.

[1274] A "match report" refers to an article that describes important scenes or events in a sports competition in written form.

[1275] "Spectator report" refers to a report that includes detailed analytical information about a sports competition.

[1276] This invention combines a system that automatically generates and synthesizes commentary, commentary, and cheering sounds for video data of sports events filmed by users with an emotion engine that recognizes the user's emotions. This system can provide a more personalized viewing experience based on the user's emotions. The system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes the emotion engine.

[1277] User terminal

[1278] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. Users select the video data using a dedicated application or web interface and press the upload button to send the data to the server.

[1279] server

[1280] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques and pattern recognition algorithms (e.g., OpenCV, TensorFlow), and records the event timestamps.

[1281] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT-4) is used to generate realistic commentary and commentary. The generated audio is then converted to a realistic form using a speech generation module (e.g., Amazon Polly, Google Text-to-Speech).

[1282] The server then uses a video editing module (e.g., FFmpeg) to properly place commentary, commentary, and cheering audio into the video, enhancing the sense of realism.

[1283] Emotion Engine

[1284] The emotion engine is a module for recognizing the user's emotions, and identifies the user's emotional state by analyzing the user's facial expressions, tone of voice, touch input, etc. This recognition uses facial expression analysis technology (e.g., dlib), voice tone analysis technology, and touch input analysis technology. Based on the emotional state recognized by the emotion engine, the server appropriately adjusts the commentary, narration, and cheering sounds it generates.

[1285] Automatic generation of match articles and match reports

[1286] The server automatically generates match articles and match reports based on the identified sporting event information, using natural language generation AI to create detailed descriptions of match highlights and important moments.

[1287] Specific examples

[1288] For example, suppose a user films a child's soccer game and uploads the video. The server analyzes the video data and detects specific events such as goals and fouls. When a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narration will become more excited.

[1289] Furthermore, based on the analysis results of the entire match, match articles such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game" are automatically generated.

[1290] Prompt Sentence Examples

[1291] Examples of prompts to be input to a generative AI model include:

[1292] "We analyzed a video of a children's soccer game. There was a goal. Please generate audio commentary such as 'Great shot!' and cheers for the goal. Also, since the user's emotions are very happy, please adjust the tone of the commentary to be more excited."

[1293] The system allows users to receive professional commentary and commentary, detailed match coverage, and a personalized viewing experience tailored to their emotions.

[1294] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1295] Step 1:

[1296] Upload video data from the user's device

[1297] Description: Users access a dedicated application or web interface using their own devices such as smartphones or PCs. After selecting the video data and pressing the upload button, the video data is sent to the server.

[1298] Input: Video data file

[1299] Output: Notification of completion of video data transmission to the server

[1300] Specific operation: The user taps the "Upload Video" button in the app, selects a video file from the device's storage, and then presses the "Upload" button to send the file to the server.

[1301] Step 2:

[1302] Video data analysis by the server

[1303] Description: The server analyzes the received video data using computer vision techniques and pattern recognition algorithms to identify specific sporting events (e.g. goals and fouls) and record them with timestamps.

[1304] Input: Uploaded video data

[1305] Output: Sport event identification results and timestamps

[1306] How it works: Once the video data arrives at the server, the video analysis process begins automatically. The server analyzes each frame, tracking player movements and ball position to detect events like goals and fouls. Detected events are then stored in a database with a timestamp.

[1307] Step 3:

[1308] Server-based generation of commentary, commentary, and cheering sounds

[1309] Description: Based on the identified event information, the server generates corresponding commentary, commentary, and cheering audio using a generative AI model (e.g., GPT-4), and uses speech synthesis technology (e.g., Amazon Polly, Google Text-to-Speech) to create an audio file from the generated text.

[1310] Input: Sport event identification results, timestamp

[1311] Output: Translated commentary and cheers text and audio files

[1312] Specific operation: For each event, the server inputs the following prompt sentence into the AI ​​model: "A goal scene has been detected. Please generate commentary text that says, 'Great shot!'" The generated text is then passed to a speech synthesis library to create an audio file.

[1313] Step 4:

[1314] Server-based audio and video data synthesis

[1315] Description: The server synthesizes the generated audio file into the video data. Using a video editing module (e.g., FFmpeg), it places the commentary, commentary, and cheering audio appropriately into the video.

[1316] Input: Original video data, generated audio file, timestamp

[1317] Output: Composite video file

[1318] How it works: Once the speech synthesis is complete, the server uses FFmpeg to add the audio file to the original video based on the timestamp. For example, a commentary saying "Great shot!" could be placed at the timestamp of a goal, creating a more immersive video.

[1319] Step 5:

[1320] Emotion recognition and feedback for users using an emotion engine

[1321] Description: The emotion engine analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state, and adjusts the tone of the commentary or commentary it generates based on the recognized emotional state.

[1322] Input: User facial expression data, tone of voice data, touch input data

[1323] Output: User's emotional state, adjusted commentary text

[1324] How it works: The emotion engine analyzes the user's previous viewing behavior and feedback. For example, if the user is very happy, the next commentary generated will be set to a higher level of excitement.

[1325] Step 6:

[1326] Automatic generation of match articles and match reports

[1327] Description: The server uses natural language generation AI to automatically generate match articles and match reports based on the analysis results and identified event information.

[1328] Input: Sporting event identification results

[1329] Output: Automatically generated match articles and match reports

[1330] What it does: The server inputs details of a sporting event into the AI ​​model and generates "match highlights and articles," such as "Player A scored a spectacular goal in the fifth minute of the first half, deciding the game."

[1331] Step 7:

[1332] Delivery of results to user devices

[1333] Description: The server delivers the generated video and match reports to the user's device, allowing the user to enjoy professional commentary, commentary, and detailed reports.

[1334] Input: Composite video files, automatically generated match reports and game reports

[1335] Output: Notification and download link to user's device

[1336] What happens: The server sends the created video file and article to the user via the app or web interface. The user logs in to download or watch the content. The app displays a notification saying "New video available," which the user can tap to go to the viewing screen.

[1337] The system provides users with a personalized and professional viewing experience.

[1338] (Application example 2)

[1339] 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."

[1340] Conventional sports viewing systems lack the means to automatically generate and synthesize realistic commentary, commentary, and cheering sounds for user-recorded video, resulting in a less than satisfactory viewing experience. It is also difficult to provide personalized content that reflects the user's emotions. This can result in a decrease in the viewer's sense of immersion and satisfaction.

[1341] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving video data of a sporting event from a user terminal, means for analyzing the received video data and identifying a specific sporting event, means for generating commentary, narration, and cheering sounds for the identified sporting event, means for recognizing the user's emotion and adjusting the commentary, narration, and cheering sounds based on the emotion, means for synthesizing the generated commentary, narration, and cheering sounds with the video data, and means for providing the synthesized video data to the user terminal. This allows the user to enjoy video with professional commentary and narration, and a viewing experience personalized according to their emotion.

[1342] "User terminal" refers to digital devices such as smartphones, tablets, and personal computers operated by users.

[1343] "Video Data" refers to data in digital format that includes footage of a sports competition.

[1344] A "sporting event" refers to a specific significant moment or occurrence in a sporting competition, such as a goal or a foul.

[1345] "Live commentary" refers to audio commentary that explains and reports the situation and plays during a sports competition in real time.

[1346] "Commentary" refers to audio commentary that includes detailed explanations of the content, tactics, and player movements of a sports competition.

[1347] "Cheer sounds" refers to sounds that simulate the cheers of spectators, expressing the excitement and thrill of moments at sporting events.

[1348] "Means for recognizing emotions" refers to a system that analyzes and identifies a user's emotional state from facial images, tone of voice, touch input, etc.

[1349] "Means for adjusting commentary, commentary, and cheering sounds" refers to a system that changes the content and tone of the commentary, commentary, and cheering sounds generated based on the user's emotional state.

[1350] "Means for synthesis" refers to the technology for overlaying the generated commentary, commentary and cheering sounds onto the original video data.

[1351] This invention is a system that automatically generates and synthesizes commentary, narration, and cheering sounds for video data of sports events by operating multiple modules in coordination, providing users with a personalized viewing experience. This system consists of a user terminal, a server, and an analysis, generation, and synthesis module that includes an emotion engine.

[1352] User terminal

[1353] The user terminal is a digital device such as a smartphone, tablet, or PC, and users use it to upload video data of sports events they have filmed to the server. The user selects the video data using a dedicated application and operates the upload button, which sends the data to the server.

[1354] server

[1355] The server analyzes the video data received from the user device and automatically identifies specific sporting events (e.g., goals, fouls, etc.) using computer vision techniques (e.g., OpenCV) and pattern recognition algorithms, and records the event timestamps.

[1356] Once the analysis is complete, the server generates corresponding commentary, commentary, and cheering audio based on the identified events. Natural language generation AI (e.g., GPT model) is used to generate realistic commentary and commentary. The generated audio is then converted into a realistic form by the speech generation module.

[1357] The server then uses a video editing module to synthesize the generated audio files into the original video data, adding a sense of realism by properly positioning the commentary, commentary, and cheering sounds into the video.

[1358] Emotion Engine

[1359] The emotion engine is a module for recognizing the user's emotions. It analyzes the user's facial expressions, tone of voice, touch input, etc. to identify their emotional state. Based on the emotional state recognized by the emotion engine, the server adjusts the generated commentary, narration, and cheering audio accordingly. This process uses the Emotion Recognizer library (e.g., EmotionRecognizer).

[1360] Automatic generation of match articles and match reports

[1361] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used to generate text that describes the highlights and important scenes of the match in detail. The generated match articles and spectator reports are also provided to the user's device.

[1362] Specific examples

[1363] For example, when a user uploads a video of their child's soccer game, the server analyzes the video and detects specific events (such as goals or fouls). If a goal is detected, a corresponding commentary such as "Great shot!" and cheers are generated and superimposed onto the original video. If the emotion engine recognizes that the user's facial expression indicates great joy, the tone of the commentary and narrator will become more excited. Similarly, based on the analysis of the entire game, a "match article" is automatically generated, including content such as "Player A scored a brilliant goal in the fifth minute of the first half, deciding the game."

[1364] Prompt Sentence Examples

[1365] An example of a prompt for a generative AI model is, "The user is very excited. Please commentate by saying, 'Great shot!' when a goal is scored."

[1366] This allows users to enjoy professional commentary and commentary, as well as a personalized viewing experience tailored to their emotions.

[1367] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1368] Step 1:

[1369] Uploading video data from user devices

[1370] Users use digital devices such as smartphones, tablets, and PCs to upload video data of sports events they have taken to a server via a dedicated application. The input is the video file taken by the user, and the output is the video file being transferred to the server.

[1371] Step 2:

[1372] Video data analysis

[1373] The server analyzes the received video data and identifies specific sporting events. This analysis uses computer vision techniques such as OpenCV and pattern recognition algorithms. The input is the uploaded video file, and the output is the timestamp information of the identified sporting events. Specifically, it sequentially scans each frame of the video data to detect characteristic events such as goals and fouls.

[1374] Step 3:

[1375] User Emotion Recognition

[1376] The server analyzes the user's facial expressions and tone of voice to recognize the user's emotions. The EmotionRecognizer library is used here. The input is the user's facial image and voice data, and the output is the recognized emotional state (e.g., joy, excitement). Specifically, it extracts features from the user's facial image and inputs them into a model to estimate emotions.

[1377] Step 4:

[1378] Generation of commentary, commentary and cheering sounds

[1379] The server generates commentary and cheering audio based on the identified sporting event and the user's emotional state. This process utilizes natural language generation AI (e.g., GPT model). The input is the timestamp information of the sporting event and the user's emotional state, and the output is a specific audio file. Specifically, the template sentence corresponding to the identified event is used as the prompt sentence, and the generative AI model generates natural commentary and commentary, which is then converted into audio.

[1380] Step 5:

[1381] Synthesizing live commentary, commentary, and cheering sounds into video data

[1382] The server synthesizes the generated commentary, commentary, and cheering audio into the original video data. This is done using a video editing module. The input is the original video file and the generated audio file, and the output is the synthesized video data. Specifically, the audio file is inserted and synchronized at the appropriate position in the video file according to the timestamp.

[1383] Step 6:

[1384] Providing the synthesized video data to the user's device

[1385] The server distributes the synthesized video data to the user terminal. The input is the synthesized video file, and the output is a video file encoded in a format that can be played on the user terminal. Specifically, the video data is provided to the user terminal in streaming or download format.

[1386] Step 7:

[1387] Automatic generation of match articles and match reports

[1388] The server automatically generates match articles and spectator reports based on the identified sporting event information. Natural language generation AI (e.g., GPT model) is used here. The input is the timestamp information and related detailed data of the sporting event, and the output is the generated text report. Specifically, it generates prompt sentences based on the highlights of each event, and then uses the natural language generation AI to create detailed sentences.

[1389] Prompt Sentence Examples

[1390] The prompt sentence is, "The user is very excited. Please commentate when the goal is scored, saying, 'Great shot!'"

[1391] 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.

[1392] 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.

[1393] 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.

[1394] 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.

[1395] 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.

[1396] 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.

[1397] 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).

[1398] 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.

[1399] 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."

[1400] 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.

[1401] 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).

[1402] 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.

[1403] 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.

[1404] 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.

[1405] 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.

[1406] 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.

[1407] 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.

[1408] 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.

[1409] 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.

[1410] 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.

[1411] 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.

[1412] The following is further disclosed regarding the above embodiment.

[1413] (Claim 1)

[1414] means for receiving video data of a sports event from a user terminal;

[1415] means for analyzing the received video data to identify a particular sporting event;

[1416] means for generating commentary and cheering sounds for the identified sporting event;

[1417] a means for synthesizing the generated commentary, commentary, and cheering sounds with video data;

[1418] means for providing the synthesized video data to a user terminal;

[1419] A system including:

[1420] (Claim 2)

[1421] 10. The system of claim 1, further comprising means for automatically generating game coverage and game reports based on information about the identified sporting events.

[1422] (Claim 3)

[1423] 10. The system of claim 1, further comprising means for providing the generated match coverage and game report to a user terminal.

[1424] "Example 1"

[1425] (Claim 1)

[1426] means for receiving video data of a sports event from a user terminal;

[1427] means for analyzing the received video data using computer vision techniques and pattern recognition algorithms to identify a particular sporting event;

[1428] a means for generating commentary and cheering sounds for the identified sporting event using natural language generation;

[1429] a means for converting the generated commentary, commentary, and cheering sounds into an audio file using an audio generation system and synthesizing the audio file with the video data;

[1430] means for providing the synthesized video data to a user terminal;

[1431] A system including:

[1432] (Claim 2)

[1433] 10. The system of claim 1, further comprising means for automatically generating game coverage and game reports using natural language generation based on the identified sporting event information.

[1434] (Claim 3)

[1435] 10. The system of claim 1, further comprising means for providing the generated match coverage and game report to a user terminal.

[1436] "Application Example 1"

[1437] (Claim 1)

[1438] means for receiving video data of a sports event from a user terminal;

[1439] means for analyzing the received video data to identify a particular sporting event;

[1440] means for generating commentary and cheering sounds for the identified sporting event;

[1441] a means for synthesizing the generated commentary, commentary, and cheering sounds with video data;

[1442] A means to share the generated video data, match articles and match reports on external platforms such as social media,

[1443] A system including:

[1444] (Claim 2)

[1445] 10. The system of claim 1, further comprising means for automatically generating game coverage and game reports based on information about the identified sporting events.

[1446] (Claim 3)

[1447] 10. The system of claim 1, further comprising means for providing the generated match coverage and game report to a user terminal.

[1448] "Example 2: Combining Emotion Engines"

[1449] (Claim 1)

[1450] means for receiving video data of a sports event from a user terminal;

[1451] means for analyzing the received video data to identify a particular sporting event;

[1452] means for generating commentary and cheering sounds for the identified sporting event;

[1453] a means for synthesizing the generated commentary, commentary, and cheering sounds with video data;

[1454] emotion recognition means for recognizing an emotion of a user;

[1455] means for adjusting the commentary or commentary generated based on the emotional state of the user;

[1456] means for providing the synthesized video data to a user terminal;

[1457] A system including:

[1458] (Claim 2)

[1459] 10. The system of claim 1, further comprising means for automatically generating game coverage and game reports based on information about the identified sporting events.

[1460] (Claim 3)

[1461] 10. The system of claim 1, further comprising means for providing the generated match coverage and game report to a user terminal.

[1462] "Application example 2 when combining emotion engines"

[1463] (Claim 1)

[1464] means for receiving video data of a sports event from a user terminal;

[1465] means for analyzing the received video data to identify a particular sporting event;

[1466] means for generating commentary and cheering sounds for the identified sporting event;

[1467] means for recognizing a user's emotion and adjusting commentary and cheering sounds based on the emotion;

[1468] a means for synthesizing the generated commentary, commentary, and cheering sounds with video data;

[1469] means for providing the synthesized video data to a user terminal;

[1470] A system including:

[1471] (Claim 2)

[1472] 10. The system of claim 1, further comprising means for automatically generating game coverage and game reports based on information about the identified sporting events.

[1473] (Claim 3)

[1474] 10. The system of claim 1, further comprising means for providing the generated match coverage and game report to a user terminal. [Explanation of symbols]

[1475] 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. means for receiving video data of a sports event from a user terminal; means for analyzing the received video data to identify a particular sporting event; means for generating commentary and cheering sounds for the identified sporting event; a means for synthesizing the generated commentary, commentary, and cheering sounds with video data; means for providing the synthesized video data to a user terminal; A system including:

2. The system of claim 1 , further comprising means for automatically generating game coverage and game reports based on the identified sporting event information.

3. The system of claim 1 , further comprising means for providing the generated match coverage and game report to a user terminal.

Images