Whisper: Courtside Edition Enhancing ASR Performance Through LLM-Driven Context Generation
This work addresses domain adaptation for ASR in sports commentary, offering a practical alternative to fine-tuning, though it is incremental as it builds on existing Whisper models.
The paper tackled the challenge of domain-specific speech recognition for NBA basketball commentary by introducing Whisper: Courtside Edition, a multi-agent LLM pipeline that enhances Whisper transcriptions without retraining, achieving a 17.0% relative reduction in word error rate from 0.217 to 0.180.
Domain-specific speech remains a persistent challenge for automatic speech recognition (ASR), even for state-of-the-art systems like OpenAI's Whisper. We introduce Whisper: Courtside Edition, a novel multi-agent large language model (LLM) pipeline that enhances Whisper transcriptions without retraining. The pipeline intercepts Whisper's initial transcript, applies specialized LLM agents for domain context identification, named entity recognition, and jargon detection, and generates compact prompts that guide Whisper's decoder. Evaluated on 421 NBA basketball commentary segments (a domain characterized by dense proper nouns and technical terminology) our best pipeline achieves a statistically significant 17.0% relative reduction in word error rate (WER; from 0.217 to 0.180, p<0.001). Improvements are observed in 40.1% of segments with degradation in only 7.1%, substantially outperforming direct transcript post-editing. These results demonstrate that prompt-based augmentation can deliver scalable domain adaptation for ASR, offering a practical alternative to costly model fine-tuning.