Text Injection for Neural Contextual Biasing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910480713383936 |
|---|---|
| author | Meng, Zhong Wu, Zelin Prabhavalkar, Rohit Peyser, Cal Wang, Weiran Chen, Nanxin Sainath, Tara N. Ramabhadran, Bhuvana |
| author_facet | Meng, Zhong Wu, Zelin Prabhavalkar, Rohit Peyser, Cal Wang, Weiran Chen, Nanxin Sainath, Tara N. Ramabhadran, Bhuvana |
| contents | Neural contextual biasing effectively improves automatic speech recognition (ASR) for crucial phrases within a speaker's context, particularly those that are infrequent in the training data. This work proposes contextual text injection (CTI) to enhance contextual ASR. CTI leverages not only the paired speech-text data, but also a much larger corpus of unpaired text to optimize the ASR model and its biasing component. Unpaired text is converted into speech-like representations and used to guide the model's attention towards relevant bias phrases. Moreover, we introduce a contextual text-injected (CTI) minimum word error rate (MWER) training, which minimizes the expected WER caused by contextual biasing when unpaired text is injected into the model. Experiments show that CTI with 100 billion text sentences can achieve up to 43.3% relative WER reduction from a strong neural biasing model. CTI-MWER provides a further relative improvement of 23.5%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_02921 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Text Injection for Neural Contextual Biasing Meng, Zhong Wu, Zelin Prabhavalkar, Rohit Peyser, Cal Wang, Weiran Chen, Nanxin Sainath, Tara N. Ramabhadran, Bhuvana Computation and Language Artificial Intelligence Machine Learning Neural and Evolutionary Computing Audio and Speech Processing Neural contextual biasing effectively improves automatic speech recognition (ASR) for crucial phrases within a speaker's context, particularly those that are infrequent in the training data. This work proposes contextual text injection (CTI) to enhance contextual ASR. CTI leverages not only the paired speech-text data, but also a much larger corpus of unpaired text to optimize the ASR model and its biasing component. Unpaired text is converted into speech-like representations and used to guide the model's attention towards relevant bias phrases. Moreover, we introduce a contextual text-injected (CTI) minimum word error rate (MWER) training, which minimizes the expected WER caused by contextual biasing when unpaired text is injected into the model. Experiments show that CTI with 100 billion text sentences can achieve up to 43.3% relative WER reduction from a strong neural biasing model. CTI-MWER provides a further relative improvement of 23.5%. |
| title | Text Injection for Neural Contextual Biasing |
| topic | Computation and Language Artificial Intelligence Machine Learning Neural and Evolutionary Computing Audio and Speech Processing |
| url | https://arxiv.org/abs/2406.02921 |