English

Conversational Speech Recognition By Learning Conversation-level Characteristics

Sound 2022-02-18 v2 Computation and Language Audio and Speech Processing

Abstract

Conversational automatic speech recognition (ASR) is a task to recognize conversational speech including multiple speakers. Unlike sentence-level ASR, conversational ASR can naturally take advantages from specific characteristics of conversation, such as role preference and topical coherence. This paper proposes a conversational ASR model which explicitly learns conversation-level characteristics under the prevalent end-to-end neural framework. The highlights of the proposed model are twofold. First, a latent variational module (LVM) is attached to a conformer-based encoder-decoder ASR backbone to learn role preference and topical coherence. Second, a topic model is specifically adopted to bias the outputs of the decoder to words in the predicted topics. Experiments on two Mandarin conversational ASR tasks show that the proposed model achieves a maximum 12% relative character error rate (CER) reduction.

Keywords

Cite

@article{arxiv.2202.07855,
  title  = {Conversational Speech Recognition By Learning Conversation-level Characteristics},
  author = {Kun Wei and Yike Zhang and Sining Sun and Lei Xie and Long Ma},
  journal= {arXiv preprint arXiv:2202.07855},
  year   = {2022}
}

Comments

ICASSP 2022

R2 v1 2026-06-24T09:40:16.312Z