Golf-Specific LLM Orchestration for Multimodal Event Insights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions are unable to generate accurate information relating to sports events, particularly in golf, using generative AI techniques that focus on mapping between text, image, and audio modalities.
Innovation Solution
A multimodal sports learning language model (LLM) processes user inputs such as text, audio, and video to determine contextual and intentional information, mapping metadata to generate sports tracking data, which is used by agents to perform actions like generating highlights, narratives, or simulating player movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generative AI techniques are used to map between text, image, and audio modalities, then the system can process multiple types of user inputs, but the accuracy of generating sports event information is insufficient
Solution Approach 1:
The system segments the generative AI model into sport-specific components (golf-specific LLM, basketball-specific LLM, etc.), each trained on domain-specific data. This segmentation allows the system to maintain high accuracy for sports event information by using specialized models while still supporting multiple input modalities through a unified architecture.
Solution Approach 2:
The patent introduces sport-specific language models as intermediary layers between the multimodal input processing and the final information generation. These intermediaries translate various input modalities into sport-specific contextual understanding, thereby improving accuracy without sacrificing versatility.
2Reliability
If a superior orchestrator selects sport-specific orchestrators based on contextual and intentional information, then the system can provide specialized sports analysis, but the system complexity increases
Solution Approach 1:
The orchestrator system is segmented into a superior orchestrator and multiple sport-specific orchestrators. Each sport-specific orchestrator handles particular sports domains, reducing the complexity burden on any single component while maintaining high reliability through specialized processing.
Solution Approach 2:
The superior orchestrator performs preliminary classification of user inputs to determine contextual and intentional information before routing to appropriate sport-specific orchestrators. This preliminary action simplifies the overall system structure by organizing complexity in a hierarchical manner.
3Measurement precision
If agents retrieve and process sports tracking data and event data, then the system can generate accurate sports content, but the data processing time increases
Solution Approach 1:
The system performs preliminary retrieval and organization of sports tracking data and event data before the actual content generation process. By preparing data in advance and organizing it according to sport-specific structures, the system reduces processing time during content generation while maintaining accuracy.
Solution Approach 2:
Sport-specific agents are designed to autonomously retrieve and process their own required data from standardized data sources. This self-service capability reduces coordination overhead and processing time while maintaining high accuracy through domain-specific data handling.
Data Source
AI summary
Disclosed techniques relate to using one or more of golf match statistics, textual insights, predictions (e.g., team and player at the event level, and team at the season level), graphics, video overlays, and player and ball tracking data. Tracking data may be generated using an in-venue feed or a broadcast feed. The tracking data may be supplemented with event data which may be provided by an operator or an automated system based on the events related to a given sport within a venue or via a broadcast feed. The tracking data and/or event data may be used to generate insights such as match statistics, textual insights, predictions, graphics, video overlays, and or the like. Accordingly, the tracking data and insights generated in accordance with the subject matter disclosed herein may be specific to a given sporting event and/or the sport associated with the sporting event.