Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915738332168192 |
|---|---|
| author | Sirichotedumrong, Warit Na-Thalang, Adisai Manakul, Potsawee Taveekitworachai, Pittawat Sripaisarnmongkol, Sittipong Pipatanakul, Kunat |
| author_facet | Sirichotedumrong, Warit Na-Thalang, Adisai Manakul, Potsawee Taveekitworachai, Pittawat Sripaisarnmongkol, Sittipong Pipatanakul, Kunat |
| contents | Large encoder-decoder models like Whisper achieve strong offline transcription but remain impractical for streaming applications due to high latency. However, due to the accessibility of pre-trained checkpoints, the open Thai ASR landscape remains dominated by these offline architectures, leaving a critical gap in efficient streaming solutions. We present Typhoon ASR Real-time, a 115M-parameter FastConformer-Transducer model for low-latency Thai speech recognition. We demonstrate that rigorous text normalization can match the impact of model scaling: our compact model achieves a 45x reduction in computational cost compared to Whisper Large-v3 while delivering comparable accuracy. Our normalization pipeline resolves systemic ambiguities in Thai transcription --including context-dependent number verbalization and repetition markers (mai yamok) --creating consistent training targets. We further introduce a two-stage curriculum learning approach for Isan (north-eastern) dialect adaptation that preserves Central Thai performance. To address reproducibility challenges in Thai ASR, we release the Typhoon ASR Benchmark, a gold-standard human-labeled datasets with transcriptions following established Thai linguistic conventions, providing standardized evaluation protocols for the research community. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_13044 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition Sirichotedumrong, Warit Na-Thalang, Adisai Manakul, Potsawee Taveekitworachai, Pittawat Sripaisarnmongkol, Sittipong Pipatanakul, Kunat Computation and Language Large encoder-decoder models like Whisper achieve strong offline transcription but remain impractical for streaming applications due to high latency. However, due to the accessibility of pre-trained checkpoints, the open Thai ASR landscape remains dominated by these offline architectures, leaving a critical gap in efficient streaming solutions. We present Typhoon ASR Real-time, a 115M-parameter FastConformer-Transducer model for low-latency Thai speech recognition. We demonstrate that rigorous text normalization can match the impact of model scaling: our compact model achieves a 45x reduction in computational cost compared to Whisper Large-v3 while delivering comparable accuracy. Our normalization pipeline resolves systemic ambiguities in Thai transcription --including context-dependent number verbalization and repetition markers (mai yamok) --creating consistent training targets. We further introduce a two-stage curriculum learning approach for Isan (north-eastern) dialect adaptation that preserves Central Thai performance. To address reproducibility challenges in Thai ASR, we release the Typhoon ASR Benchmark, a gold-standard human-labeled datasets with transcriptions following established Thai linguistic conventions, providing standardized evaluation protocols for the research community. |
| title | Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.13044 |