Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2508.15853 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915456540999680 |
|---|---|
| author | Yang, Xuwen |
| author_facet | Yang, Xuwen |
| contents | End-to-end ASR models, despite their success on benchmarks, often pro-duce catastrophic semantic errors in noisy environments. We attribute this fragility to the prevailing 'direct mapping' objective, which solely penalizes final output errors while leaving the model's internal computational pro-cess unconstrained. To address this, we introduce the Multi-Granularity Soft Consistency (MGSC) framework, a model-agnostic, plug-and-play module that enforces internal self-consistency by simultaneously regulariz-ing macro-level sentence semantics and micro-level token alignment. Cru-cially, our work is the first to uncover a powerful synergy between these two consistency granularities: their joint optimization yields robustness gains that significantly surpass the sum of their individual contributions. On a public dataset, MGSC reduces the average Character Error Rate by a relative 8.7% across diverse noise conditions, primarily by preventing se-vere meaning-altering mistakes. Our work demonstrates that enforcing in-ternal consistency is a crucial step towards building more robust and trust-worthy AI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_15853 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MGSC: A Multi-granularity Consistency Framework for Robust End-to-end Asr Yang, Xuwen Computation and Language Artificial Intelligence Sound Audio and Speech Processing I.2.7 End-to-end ASR models, despite their success on benchmarks, often pro-duce catastrophic semantic errors in noisy environments. We attribute this fragility to the prevailing 'direct mapping' objective, which solely penalizes final output errors while leaving the model's internal computational pro-cess unconstrained. To address this, we introduce the Multi-Granularity Soft Consistency (MGSC) framework, a model-agnostic, plug-and-play module that enforces internal self-consistency by simultaneously regulariz-ing macro-level sentence semantics and micro-level token alignment. Cru-cially, our work is the first to uncover a powerful synergy between these two consistency granularities: their joint optimization yields robustness gains that significantly surpass the sum of their individual contributions. On a public dataset, MGSC reduces the average Character Error Rate by a relative 8.7% across diverse noise conditions, primarily by preventing se-vere meaning-altering mistakes. Our work demonstrates that enforcing in-ternal consistency is a crucial step towards building more robust and trust-worthy AI. |
| title | MGSC: A Multi-granularity Consistency Framework for Robust End-to-end Asr |
| topic | Computation and Language Artificial Intelligence Sound Audio and Speech Processing I.2.7 |
| url | https://arxiv.org/abs/2508.15853 |