English

Contextualized End-to-End Speech Recognition with Contextual Phrase Prediction Network

Audio and Speech Processing 2023-07-13 v5 Computation and Language Sound

Abstract

Contextual information plays a crucial role in speech recognition technologies and incorporating it into the end-to-end speech recognition models has drawn immense interest recently. However, previous deep bias methods lacked explicit supervision for bias tasks. In this study, we introduce a contextual phrase prediction network for an attention-based deep bias method. This network predicts context phrases in utterances using contextual embeddings and calculates bias loss to assist in the training of the contextualized model. Our method achieved a significant word error rate (WER) reduction across various end-to-end speech recognition models. Experiments on the LibriSpeech corpus show that our proposed model obtains a 12.1% relative WER improvement over the baseline model, and the WER of the context phrases decreases relatively by 40.5%. Moreover, by applying a context phrase filtering strategy, we also effectively eliminate the WER degradation when using a larger biasing list.

Keywords

Cite

@article{arxiv.2305.12493,
  title  = {Contextualized End-to-End Speech Recognition with Contextual Phrase Prediction Network},
  author = {Kaixun Huang and Ao Zhang and Zhanheng Yang and Pengcheng Guo and Bingshen Mu and Tianyi Xu and Lei Xie},
  journal= {arXiv preprint arXiv:2305.12493},
  year   = {2023}
}

Comments

Accepted by interspeech2023

R2 v1 2026-06-28T10:40:33.722Z