Iterative Pseudo-Labeling for Speech Recognition

TL;DR

This paper introduces an iterative pseudo-labeling method for speech recognition that improves model accuracy by leveraging unlabeled data through multiple training cycles, achieving state-of-the-art results.

Contribution

The paper presents a novel iterative pseudo-labeling approach that enhances semi-supervised speech recognition by effectively utilizing unlabeled data and language models.

Findings

Achieved state-of-the-art word-error rates on Librispeech test sets.

Demonstrated effective use of additional text data with IPL.

Showed IPL's robustness in low-resource settings.

Abstract

Pseudo-labeling has recently shown promise in end-to-end automatic speech recognition (ASR). We study Iterative Pseudo-Labeling (IPL), a semi-supervised algorithm which efficiently performs multiple iterations of pseudo-labeling on unlabeled data as the acoustic model evolves. In particular, IPL fine-tunes an existing model at each iteration using both labeled data and a subset of unlabeled data. We study the main components of IPL: decoding with a language model and data augmentation. We then demonstrate the effectiveness of IPL by achieving state-of-the-art word-error rate on the Librispeech test sets in both standard and low-resource setting. We also study the effect of language models trained on different corpora to show IPL can effectively utilize additional text. Finally, we release a new large in-domain text corpus which does not overlap with the Librispeech training transcriptions to foster research in low-resource, semi-supervised ASR