Towards Robust Dysarthric Speech Recognition: LLM-Agent Post-ASR Correction Beyond WER
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917231225470976 |
|---|---|
| author | Zheng, Xiuwen Dong, Sixun Phukon, Bornali Hasegawa-Johnson, Mark Yoo, Chang D. |
| author_facet | Zheng, Xiuwen Dong, Sixun Phukon, Bornali Hasegawa-Johnson, Mark Yoo, Chang D. |
| contents | While Automatic Speech Recognition (ASR) is typically benchmarked by word error rate (WER), real-world applications ultimately hinge on semantic fidelity. This mismatch is particularly problematic for dysarthric speech, where articulatory imprecision and disfluencies can cause severe semantic distortions. To bridge this gap, we introduce a Large Language Model (LLM)-based agent for post-ASR correction: a Judge-Editor over the top-k ASR hypotheses that keeps high-confidence spans, rewrites uncertain segments, and operates in both zero-shot and fine-tuned modes. In parallel, we release SAP-Hypo5, the largest benchmark for dysarthric speech correction, to enable reproducibility and future exploration. Under multi-perspective evaluation, our agent achieves a 14.51% WER reduction alongside substantial semantic gains, including a +7.59 pp improvement in MENLI and +7.66 pp in Slot Micro F1 on challenging samples. Our analysis further reveals that WER is highly sensitive to domain shift, whereas semantic metrics correlate more closely with downstream task performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21347 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Towards Robust Dysarthric Speech Recognition: LLM-Agent Post-ASR Correction Beyond WER Zheng, Xiuwen Dong, Sixun Phukon, Bornali Hasegawa-Johnson, Mark Yoo, Chang D. Audio and Speech Processing Sound While Automatic Speech Recognition (ASR) is typically benchmarked by word error rate (WER), real-world applications ultimately hinge on semantic fidelity. This mismatch is particularly problematic for dysarthric speech, where articulatory imprecision and disfluencies can cause severe semantic distortions. To bridge this gap, we introduce a Large Language Model (LLM)-based agent for post-ASR correction: a Judge-Editor over the top-k ASR hypotheses that keeps high-confidence spans, rewrites uncertain segments, and operates in both zero-shot and fine-tuned modes. In parallel, we release SAP-Hypo5, the largest benchmark for dysarthric speech correction, to enable reproducibility and future exploration. Under multi-perspective evaluation, our agent achieves a 14.51% WER reduction alongside substantial semantic gains, including a +7.59 pp improvement in MENLI and +7.66 pp in Slot Micro F1 on challenging samples. Our analysis further reveals that WER is highly sensitive to domain shift, whereas semantic metrics correlate more closely with downstream task performance. |
| title | Towards Robust Dysarthric Speech Recognition: LLM-Agent Post-ASR Correction Beyond WER |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2601.21347 |