Something from Nothing: Data Augmentation for Robust Severity Level Estimation of Dysarthric Speech
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_ | 1866914523077672960 |
|---|---|
| author | Bae, Jaesung Zheng, Xiuwen Kim, Minje Yoo, Chang D. Hasegawa-Johnson, Mark |
| author_facet | Bae, Jaesung Zheng, Xiuwen Kim, Minje Yoo, Chang D. Hasegawa-Johnson, Mark |
| contents | Dysarthric speech quality assessment (DSQA) is critical for clinical diagnostics and inclusive speech technologies. However, subjective evaluation is costly and difficult to scale, and the scarcity of labeled data limits robust objective modeling. To address this, we propose a three-stage framework that leverages unlabeled dysarthric speech and large-scale typical speech datasets to scale training. A teacher model first generates pseudo-labels for unlabeled samples, followed by weakly supervised pretraining using a label-aware contrastive learning strategy that exposes the model to diverse speakers and acoustic conditions. The pretrained model is then fine-tuned for the downstream DSQA task. Experiments on five unseen datasets spanning multiple etiologies and languages demonstrate the robustness of our approach. Our Whisper-based baseline significantly outperforms SOTA DSQA predictors such as SpICE, and the full framework achieves an average SRCC of 0.761 across unseen test datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15988 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Something from Nothing: Data Augmentation for Robust Severity Level Estimation of Dysarthric Speech Bae, Jaesung Zheng, Xiuwen Kim, Minje Yoo, Chang D. Hasegawa-Johnson, Mark Audio and Speech Processing Artificial Intelligence Machine Learning Dysarthric speech quality assessment (DSQA) is critical for clinical diagnostics and inclusive speech technologies. However, subjective evaluation is costly and difficult to scale, and the scarcity of labeled data limits robust objective modeling. To address this, we propose a three-stage framework that leverages unlabeled dysarthric speech and large-scale typical speech datasets to scale training. A teacher model first generates pseudo-labels for unlabeled samples, followed by weakly supervised pretraining using a label-aware contrastive learning strategy that exposes the model to diverse speakers and acoustic conditions. The pretrained model is then fine-tuned for the downstream DSQA task. Experiments on five unseen datasets spanning multiple etiologies and languages demonstrate the robustness of our approach. Our Whisper-based baseline significantly outperforms SOTA DSQA predictors such as SpICE, and the full framework achieves an average SRCC of 0.761 across unseen test datasets. |
| title | Something from Nothing: Data Augmentation for Robust Severity Level Estimation of Dysarthric Speech |
| topic | Audio and Speech Processing Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2603.15988 |