Zero-Shot End-To-End Spoken Question Answering In Medical Domain
This work addresses resource constraints and error accumulation in medical spoken question answering, though it appears incremental as it builds on existing end-to-end methodologies.
The paper tackled the problem of resource-intensive and error-prone spoken question answering in the medical domain by proposing a zero-shot end-to-end approach, which reduced resource usage by up to 14.7 times and improved average accuracy by 0.5% compared to traditional cascade systems.
In the rapidly evolving landscape of spoken question-answering (SQA), the integration of large language models (LLMs) has emerged as a transformative development. Conventional approaches often entail the use of separate models for question audio transcription and answer selection, resulting in significant resource utilization and error accumulation. To tackle these challenges, we explore the effectiveness of end-to-end (E2E) methodologies for SQA in the medical domain. Our study introduces a novel zero-shot SQA approach, compared to traditional cascade systems. Through a comprehensive evaluation conducted on a new open benchmark of 8 medical tasks and 48 hours of synthetic audio, we demonstrate that our approach requires up to 14.7 times fewer resources than a combined 1.3B parameters LLM with a 1.55B parameters ASR model while improving average accuracy by 0.5\%. These findings underscore the potential of E2E methodologies for SQA in resource-constrained contexts.