Minimum Latency Training of Sequence Transducers for Streaming End-to-End Speech Recognition

Yusuke Shinohara, Shinji Watanabe

Research output: Contribution to journalConference articlepeer-review

Abstract

Sequence transducers, such as the RNN-T and the Conformer-T, are one of the most promising models of end-to-end speech recognition, especially in streaming scenarios where both latency and accuracy are important. Although various methods, such as alignment-restricted training and FastEmit, have been studied to reduce the latency, latency reduction is often accompanied with a significant degradation in accuracy. We argue that this suboptimal performance might be caused because none of the prior methods explicitly model and reduce the latency. In this paper, we propose a new training method to explicitly model and reduce the latency of sequence transducer models. First, we define the expected latency at each diagonal line on the lattice, and show that its gradient can be computed efficiently within the forward-backward algorithm. Then we augment the transducer loss with this expected latency, so that an optimal trade-off between latency and accuracy is achieved. Experimental results on the WSJ dataset show that the proposed minimum latency training reduces the latency of causal Conformer-T from 220 ms to 27 ms within a WER degradation of 0.7%, and outperforms conventional alignment-restricted training (110 ms) and FastEmit (67 ms) methods.

Original languageEnglish
Pages (from-to)2098-2102
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Volume2022-September
DOIs
Publication statusPublished - 2022
Externally publishedYes
Event23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 - Incheon, Korea, Republic of
Duration: 2022 Sep 182022 Sep 22

Keywords

  • end-to-end
  • latency
  • sequence transducer
  • speech recognition
  • streaming

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'Minimum Latency Training of Sequence Transducers for Streaming End-to-End Speech Recognition'. Together they form a unique fingerprint.

Cite this