Spike-Triggered Contextual Biasing for End-to-End Mandarin Speech Recognition

TL;DR

This paper introduces a spike-triggered deep biasing method for Mandarin speech recognition that effectively combines explicit and implicit contextual biasing, significantly reducing error rates especially on biased phrases.

Contribution

It proposes a novel spike-triggered biasing approach supporting both explicit and implicit bias, enhancing recognition accuracy in end-to-end Mandarin ASR systems.

Findings

32.0% relative CER reduction overall

68.6% relative CER reduction on contextual phrases

Effective combination with shallow fusion methods

Abstract

The attention-based deep contextual biasing method has been demonstrated to effectively improve the recognition performance of end-to-end automatic speech recognition (ASR) systems on given contextual phrases. However, unlike shallow fusion methods that directly bias the posterior of the ASR model, deep biasing methods implicitly integrate contextual information, making it challenging to control the degree of bias. In this study, we introduce a spike-triggered deep biasing method that simultaneously supports both explicit and implicit bias. Moreover, both bias approaches exhibit significant improvements and can be cascaded with shallow fusion methods for better results. Furthermore, we propose a context sampling enhancement strategy and improve the contextual phrase filtering algorithm. Experiments on the public WenetSpeech Mandarin biased-word dataset show a 32.0% relative CER reduction compared to the baseline model, with an impressively 68.6% relative CER reduction on contextual phrases.