Unsupervised Online Continual Learning for Automatic Speech Recognition

TL;DR

This paper introduces an unsupervised online continual learning approach for automatic speech recognition that reduces catastrophic forgetting and enables models to adapt to new domains using unlabeled data.

Contribution

It extends online continual learning for ASR into an unsupervised setting using self-training, significantly reducing forgetting and improving domain adaptation without labeled data.

Findings

UOCL suffers less forgetting than supervised OCL

UOCL approaches the performance of supervised methods

Proposed UOCL extensions further improve adaptation

Abstract

Adapting Automatic Speech Recognition (ASR) models to new domains leads to Catastrophic Forgetting (CF) of previously learned information. This paper addresses CF in the challenging context of Online Continual Learning (OCL), with tasks presented as a continuous data stream with unknown boundaries. We extend OCL for ASR into the unsupervised realm, by leveraging self-training (ST) to facilitate unsupervised adaptation, enabling models to adapt continually without label dependency and without forgetting previous knowledge. Through comparative analysis of various OCL and ST methods across two domain adaptation experiments, we show that UOCL suffers from significantly less forgetting compared to supervised OCL, allowing UOCL methods to approach the performance levels of supervised OCL. Our proposed UOCL extensions further boosts UOCL's efficacy. Our findings represent a significant step towards continually adaptable ASR systems, capable of leveraging unlabeled data across diverse domains.