Context-Aware Dynamic Chunking for Streaming Tibetan Speech Recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914153273229312 |
|---|---|
| author | Wang, Chao Cai, Yuqing Duojie, Renzeng Zhang, Jin Liu, Yutong Tashi, Nyima |
| author_facet | Wang, Chao Cai, Yuqing Duojie, Renzeng Zhang, Jin Liu, Yutong Tashi, Nyima |
| contents | In this work, we propose a streaming speech recognition framework for Amdo Tibetan, built upon a hybrid CTC/Atten-tion architecture with a context-aware dynamic chunking mechanism. The proposed strategy adaptively adjusts chunk widths based on encoding states, enabling flexible receptive fields, cross-chunk information exchange, and robust adaptation to varying speaking rates, thereby alleviating the context truncation problem of fixed-chunk methods. To further capture the linguistic characteristics of Tibetan, we construct a lexicon grounded in its orthographic principles, providing linguistically motivated modeling units. During decoding, an external language model is integrated to enhance semantic consistency and improve recognition of long sentences. Experimental results show that the proposed framework achieves a word error rate (WER) of 6.23% on the test set, yielding a 48.15% relative improvement over the fixed-chunk baseline, while significantly reducing recognition latency and maintaining performance close to global decoding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_09085 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Context-Aware Dynamic Chunking for Streaming Tibetan Speech Recognition Wang, Chao Cai, Yuqing Duojie, Renzeng Zhang, Jin Liu, Yutong Tashi, Nyima Computation and Language In this work, we propose a streaming speech recognition framework for Amdo Tibetan, built upon a hybrid CTC/Atten-tion architecture with a context-aware dynamic chunking mechanism. The proposed strategy adaptively adjusts chunk widths based on encoding states, enabling flexible receptive fields, cross-chunk information exchange, and robust adaptation to varying speaking rates, thereby alleviating the context truncation problem of fixed-chunk methods. To further capture the linguistic characteristics of Tibetan, we construct a lexicon grounded in its orthographic principles, providing linguistically motivated modeling units. During decoding, an external language model is integrated to enhance semantic consistency and improve recognition of long sentences. Experimental results show that the proposed framework achieves a word error rate (WER) of 6.23% on the test set, yielding a 48.15% relative improvement over the fixed-chunk baseline, while significantly reducing recognition latency and maintaining performance close to global decoding. |
| title | Context-Aware Dynamic Chunking for Streaming Tibetan Speech Recognition |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2511.09085 |