Improving RNN-Transducers with Acoustic LookAhead

TL;DR

This paper introduces LookAhead, a method to enhance RNN-Transducers by making text representations more acoustically grounded, resulting in significant improvements in speech-to-text accuracy across various datasets.

Contribution

We propose LookAhead, a novel technique that incorporates future audio context into text representations in RNN-T models, reducing hallucinations and improving accuracy.

Findings

5%-20% relative WER reduction on evaluation sets

Improved robustness to out-of-domain data

Enhanced acoustic grounding of text representations

Abstract

RNN-Transducers (RNN-Ts) have gained widespread acceptance as an end-to-end model for speech to text conversion because of their high accuracy and streaming capabilities. A typical RNN-T independently encodes the input audio and the text context, and combines the two encodings by a thin joint network. While this architecture provides SOTA streaming accuracy, it also makes the model vulnerable to strong LM biasing which manifests as multi-step hallucination of text without acoustic evidence. In this paper we propose LookAhead that makes text representations more acoustically grounded by looking ahead into the future within the audio input. This technique yields a significant 5%-20% relative reduction in word error rate on both in-domain and out-of-domain evaluation sets.