ViSpeechFormer: A Phonemic Approach for Vietnamese Automatic 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_ | 1866910018059632640 |
|---|---|
| author | Nguyen, Khoa Anh Hoang, Long Minh Nguyen, Nghia Hieu Nguyen, Luan Thanh Nguyen, Ngan Luu-Thuy |
| author_facet | Nguyen, Khoa Anh Hoang, Long Minh Nguyen, Nghia Hieu Nguyen, Luan Thanh Nguyen, Ngan Luu-Thuy |
| contents | Vietnamese has a phonetic orthography, where each grapheme corresponds to at most one phoneme and vice versa. Exploiting this high grapheme-phoneme transparency, we propose ViSpeechFormer (\textbf{Vi}etnamese \textbf{Speech} Trans\textbf{Former}), a phoneme-based approach for Vietnamese Automatic Speech Recognition (ASR). To the best of our knowledge, this is the first Vietnamese ASR framework that explicitly models phonemic representations. Experiments on two publicly available Vietnamese ASR datasets show that ViSpeechFormer achieves strong performance, generalizes better to out-of-vocabulary words, and is less affected by training bias. This phoneme-based paradigm is also promising for other languages with phonetic orthographies. The code will be released upon acceptance of this paper. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_10003 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ViSpeechFormer: A Phonemic Approach for Vietnamese Automatic Speech Recognition Nguyen, Khoa Anh Hoang, Long Minh Nguyen, Nghia Hieu Nguyen, Luan Thanh Nguyen, Ngan Luu-Thuy Computation and Language Vietnamese has a phonetic orthography, where each grapheme corresponds to at most one phoneme and vice versa. Exploiting this high grapheme-phoneme transparency, we propose ViSpeechFormer (\textbf{Vi}etnamese \textbf{Speech} Trans\textbf{Former}), a phoneme-based approach for Vietnamese Automatic Speech Recognition (ASR). To the best of our knowledge, this is the first Vietnamese ASR framework that explicitly models phonemic representations. Experiments on two publicly available Vietnamese ASR datasets show that ViSpeechFormer achieves strong performance, generalizes better to out-of-vocabulary words, and is less affected by training bias. This phoneme-based paradigm is also promising for other languages with phonetic orthographies. The code will be released upon acceptance of this paper. |
| title | ViSpeechFormer: A Phonemic Approach for Vietnamese Automatic Speech Recognition |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.10003 |