MADI: Inter-domain Matching and Intra-domain Discrimination for Cross-domain Speech Recognition

TL;DR

This paper introduces MADI, a novel unsupervised domain adaptation method for speech recognition that enhances transferability and discriminability, significantly reducing word error rates across different domains.

Contribution

MADI combines inter-domain matching with intra-domain discrimination to improve cross-domain speech recognition performance.

Findings

Reduces WER by 17.7% on cross-device tasks.

Reduces WER by 22.8% on cross-environment tasks.

Effective on Libri-Adapt dataset.

Abstract

End-to-end automatic speech recognition (ASR) usually suffers from performance degradation when applied to a new domain due to domain shift. Unsupervised domain adaptation (UDA) aims to improve the performance on the unlabeled target domain by transferring knowledge from the source to the target domain. To improve transferability, existing UDA approaches mainly focus on matching the distributions of the source and target domains globally and/or locally, while ignoring the model discriminability. In this paper, we propose a novel UDA approach for ASR via inter-domain MAtching and intra-domain DIscrimination (MADI), which improves the model transferability by fine-grained inter-domain matching and discriminability by intra-domain contrastive discrimination simultaneously. Evaluations on the Libri-Adapt dataset demonstrate the effectiveness of our approach. MADI reduces the relative word error rate (WER) on cross-device and cross-environment ASR by 17.7% and 22.8%, respectively.