When Audio-Language Models Fail to Leverage Multimodal Context for Dysarthric Speech Recognition

TL;DR

This paper evaluates whether current audio-language models can utilize clinical context to improve dysarthric speech recognition, finding they largely do not, but fine-tuning shows significant improvements in specific subgroups.

Contribution

The study introduces a benchmark for testing clinical context use in dysarthric speech recognition and demonstrates that fine-tuning with clinical prompts can substantially improve performance.

Findings

Models do not significantly benefit from clinical context prompts.

Fine-tuning with LoRA reduces WER by 52% compared to baseline.

Significant gains observed for Down syndrome and mild-severity speakers.

Abstract

Automatic speech recognition (ASR) systems remain brittle on dysarthric and other atypical speech. Recent audio-language models raise the possibility of improving performance by conditioning on additional clinical context at inference time, but it is unclear whether these models can make use of such information. We introduce a benchmark built on the Speech Accessibility Project (SAP) dataset that tests whether diagnosis labels, clinician-derived speech ratings, and progressively richer clinical descriptions improve transcription accuracy for dysarthric speech. Across matched comparisons on nine models, we find that current models do not meaningfully use this context: diagnosis-informed and clinically detailed prompts yield negligible improvements and often degrade word error rate. We complement the prompting analysis with context-dependent fine-tuning, showing that LoRA adaptation with a mixture of clinical prompt formats achieves a WER of 0.066, a 52% relative reduction over the frozen baseline, while preserving performance when context is unavailable. Subgroup analyses reveal significant gains for Down syndrome and mild-severity speakers. These results clarify where current models fall short and provide a testbed for measuring progress toward more inclusive ASR.