Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition
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_ | 1866917241779388416 |
|---|---|
| author | Lee, Wonjun Kim, Hyounghun Lee, Gary Geunbae |
| author_facet | Lee, Wonjun Kim, Hyounghun Lee, Gary Geunbae |
| contents | Accented speech remains a persistent challenge for automatic speech recognition (ASR), as most models are trained on data dominated by a few high-resource English varieties, leading to substantial performance degradation for other accents. Accent-agnostic approaches improve robustness yet struggle with heavily accented or unseen varieties, while accent-specific methods rely on limited and often noisy labels. We introduce Moe-Ctc, a Mixture-of-Experts architecture with intermediate CTC supervision that jointly promotes expert specialization and generalization. During training, accent-aware routing encourages experts to capture accent-specific patterns, which gradually transitions to label-free routing for inference. Each expert is equipped with its own CTC head to align routing with transcription quality, and a routing-augmented loss further stabilizes optimization. Experiments on the Mcv-Accent benchmark demonstrate consistent gains across both seen and unseen accents in low- and high-resource conditions, achieving up to 29.3% relative WER reduction over strong FastConformer baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_01967 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition Lee, Wonjun Kim, Hyounghun Lee, Gary Geunbae Computation and Language Artificial Intelligence Accented speech remains a persistent challenge for automatic speech recognition (ASR), as most models are trained on data dominated by a few high-resource English varieties, leading to substantial performance degradation for other accents. Accent-agnostic approaches improve robustness yet struggle with heavily accented or unseen varieties, while accent-specific methods rely on limited and often noisy labels. We introduce Moe-Ctc, a Mixture-of-Experts architecture with intermediate CTC supervision that jointly promotes expert specialization and generalization. During training, accent-aware routing encourages experts to capture accent-specific patterns, which gradually transitions to label-free routing for inference. Each expert is equipped with its own CTC head to align routing with transcription quality, and a routing-augmented loss further stabilizes optimization. Experiments on the Mcv-Accent benchmark demonstrate consistent gains across both seen and unseen accents in low- and high-resource conditions, achieving up to 29.3% relative WER reduction over strong FastConformer baselines. |
| title | Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2602.01967 |