DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910809936887808 |
|---|---|
| author | Lee, Wonjun Im, Solee Do, Heejin Kim, Yunsu Ok, Jungseul Lee, Gary Geunbae |
| author_facet | Lee, Wonjun Im, Solee Do, Heejin Kim, Yunsu Ok, Jungseul Lee, Gary Geunbae |
| contents | Dysarthric speech recognition often suffers from performance degradation due to the intrinsic diversity of dysarthric severity and extrinsic disparity from normal speech. To bridge these gaps, we propose a Dynamic Phoneme-level Contrastive Learning (DyPCL) method, which leads to obtaining invariant representations across diverse speakers. We decompose the speech utterance into phoneme segments for phoneme-level contrastive learning, leveraging dynamic connectionist temporal classification alignment. Unlike prior studies focusing on utterance-level embeddings, our granular learning allows discrimination of subtle parts of speech. In addition, we introduce dynamic curriculum learning, which progressively transitions from easy negative samples to difficult-to-distinguishable negative samples based on phonetic similarity of phoneme. Our approach to training by difficulty levels alleviates the inherent variability of speakers, better identifying challenging speeches. Evaluated on the UASpeech dataset, DyPCL outperforms baseline models, achieving an average 22.10\% relative reduction in word error rate (WER) across the overall dysarthria group. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_19010 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition Lee, Wonjun Im, Solee Do, Heejin Kim, Yunsu Ok, Jungseul Lee, Gary Geunbae Computation and Language Sound Audio and Speech Processing Dysarthric speech recognition often suffers from performance degradation due to the intrinsic diversity of dysarthric severity and extrinsic disparity from normal speech. To bridge these gaps, we propose a Dynamic Phoneme-level Contrastive Learning (DyPCL) method, which leads to obtaining invariant representations across diverse speakers. We decompose the speech utterance into phoneme segments for phoneme-level contrastive learning, leveraging dynamic connectionist temporal classification alignment. Unlike prior studies focusing on utterance-level embeddings, our granular learning allows discrimination of subtle parts of speech. In addition, we introduce dynamic curriculum learning, which progressively transitions from easy negative samples to difficult-to-distinguishable negative samples based on phonetic similarity of phoneme. Our approach to training by difficulty levels alleviates the inherent variability of speakers, better identifying challenging speeches. Evaluated on the UASpeech dataset, DyPCL outperforms baseline models, achieving an average 22.10\% relative reduction in word error rate (WER) across the overall dysarthria group. |
| title | DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2501.19010 |