CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912134068174848 |
|---|---|
| author | Zhang, Tian-Hao Zhou, Dinghao Zhong, Guiping Zhou, Jiaming Li, Baoxiang |
| author_facet | Zhang, Tian-Hao Zhou, Dinghao Zhong, Guiping Zhou, Jiaming Li, Baoxiang |
| contents | RNN-T models are widely used in ASR, which rely on the RNN-T loss to achieve length alignment between input audio and target sequence. However, the implementation complexity and the alignment-based optimization target of RNN-T loss lead to computational redundancy and a reduced role for predictor network, respectively. In this paper, we propose a novel model named CIF-Transducer (CIF-T) which incorporates the Continuous Integrate-and-Fire (CIF) mechanism with the RNN-T model to achieve efficient alignment. In this way, the RNN-T loss is abandoned, thus bringing a computational reduction and allowing the predictor network a more significant role. We also introduce Funnel-CIF, Context Blocks, Unified Gating and Bilinear Pooling joint network, and auxiliary training strategy to further improve performance. Experiments on the 178-hour AISHELL-1 and 10000-hour WenetSpeech datasets show that CIF-T achieves state-of-the-art results with lower computational overhead compared to RNN-T models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_14132 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition Zhang, Tian-Hao Zhou, Dinghao Zhong, Guiping Zhou, Jiaming Li, Baoxiang Sound Computation and Language Audio and Speech Processing RNN-T models are widely used in ASR, which rely on the RNN-T loss to achieve length alignment between input audio and target sequence. However, the implementation complexity and the alignment-based optimization target of RNN-T loss lead to computational redundancy and a reduced role for predictor network, respectively. In this paper, we propose a novel model named CIF-Transducer (CIF-T) which incorporates the Continuous Integrate-and-Fire (CIF) mechanism with the RNN-T model to achieve efficient alignment. In this way, the RNN-T loss is abandoned, thus bringing a computational reduction and allowing the predictor network a more significant role. We also introduce Funnel-CIF, Context Blocks, Unified Gating and Bilinear Pooling joint network, and auxiliary training strategy to further improve performance. Experiments on the 178-hour AISHELL-1 and 10000-hour WenetSpeech datasets show that CIF-T achieves state-of-the-art results with lower computational overhead compared to RNN-T models. |
| title | CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition |
| topic | Sound Computation and Language Audio and Speech Processing |
| url | https://arxiv.org/abs/2307.14132 |