BaseballMind is an AI-powered baseball broadcasting system that integrates domain-fine-tuned ASR, LLM-based intent and slot parsing, and a YOLO–OCR scoreboard state machine. It performs temporal alignment and cross-modal validation across broadcast audio, captions, and video to automatically detect key events such as home runs, strikeouts, and walks while resolving player nicknames. Structured event tagging enables natural-language or voice queries to instantly locate and retrieve relevant video clips, supporting real-time production, highlight editing, and intelligent content services.

The core innovation is a closed-loop multimodal event-tagging framework that integrates domain-fine-tuned ASR, LLM-based intent and slot parsing, and a YOLO–OCR scoreboard state machine. It performs cross-modal alignment and validation across speech, captions, and video to automatically identify event types and temporal boundaries, reducing manual annotation. Using game knowledge, the LLM supports player entity disambiguation, nickname resolution, and ASR error correction. The query layer applies controlled API routing and parameterized queries to prevent direct SQL generation while ensuring accuracy and data security.

The event-driven architecture targets three verticals: coaching teams gain rapid retrieval of situational clips for opponent scouting and player performance analysis; broadcast production teams automate highlight indexing and reduce post-production workload; fan engagement platforms enable natural language search without database expertise. The structured event schema—covering inning, count, base state, and pitcher-batter matchup—supports cross-season analytics and multi-sport portability.