USTED: Improving ASR with a Unified Speech and Text Encoder-Decoder

TL;DR

This paper introduces USTED, a unified speech and text encoder-decoder model that improves end-to-end speech recognition by jointly training with auxiliary text tasks, reducing WER without extra inference costs.

Contribution

The paper presents a novel joint training approach combining ASR with auxiliary text tasks, enhancing performance without additional inference complexity.

Findings

WER reduced by 16% and 20% on test sets

Achieves further gains with additional auxiliary tasks

No extra inference cost compared to traditional models

Abstract

Improving end-to-end speech recognition by incorporating external text data has been a longstanding research topic. There has been a recent focus on training E2E ASR models that get the performance benefits of external text data without incurring the extra cost of evaluating an external language model at inference time. In this work, we propose training ASR model jointly with a set of text-to-text auxiliary tasks with which it shares a decoder and parts of the encoder. When we jointly train ASR and masked language model with the 960-hour Librispeech and Opensubtitles data respectively, we observe WER reductions of 16% and 20% on test-other and test-clean respectively over an ASR-only baseline without any extra cost at inference time, and reductions of 6% and 8% compared to a stronger MUTE-L baseline which trains the decoder with the same text data as our model. We achieve further improvements when we train masked language model on Librispeech data or when we use machine translation as the auxiliary task, without significantly sacrificing performance on the task itself.