Paraformer-v2: An improved non-autoregressive transformer for noise-robust speech recognition
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914957959888896 |
|---|---|
| author | An, Keyu Li, Zerui Gao, Zhifu Zhang, Shiliang |
| author_facet | An, Keyu Li, Zerui Gao, Zhifu Zhang, Shiliang |
| contents | Attention-based encoder-decoder, e.g. transformer and its variants, generates the output sequence in an autoregressive (AR) manner. Despite its superior performance, AR model is computationally inefficient as its generation requires as many iterations as the output length. In this paper, we propose Paraformer-v2, an improved version of Paraformer, for fast, accurate, and noise-robust non-autoregressive speech recognition. In Paraformer-v2, we use a CTC module to extract the token embeddings, as the alternative to the continuous integrate-and-fire module in Paraformer. Extensive experiments demonstrate that Paraformer-v2 outperforms Paraformer on multiple datasets, especially on the English datasets (over 14% improvement on WER), and is more robust in noisy environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_17746 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Paraformer-v2: An improved non-autoregressive transformer for noise-robust speech recognition An, Keyu Li, Zerui Gao, Zhifu Zhang, Shiliang Audio and Speech Processing Sound Attention-based encoder-decoder, e.g. transformer and its variants, generates the output sequence in an autoregressive (AR) manner. Despite its superior performance, AR model is computationally inefficient as its generation requires as many iterations as the output length. In this paper, we propose Paraformer-v2, an improved version of Paraformer, for fast, accurate, and noise-robust non-autoregressive speech recognition. In Paraformer-v2, we use a CTC module to extract the token embeddings, as the alternative to the continuous integrate-and-fire module in Paraformer. Extensive experiments demonstrate that Paraformer-v2 outperforms Paraformer on multiple datasets, especially on the English datasets (over 14% improvement on WER), and is more robust in noisy environments. |
| title | Paraformer-v2: An improved non-autoregressive transformer for noise-robust speech recognition |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2409.17746 |