Make More of Your Data: Minimal Effort Data Augmentation for Automatic Speech Recognition and Translation

TL;DR

This paper presents a simple, cost-effective data augmentation method for speech recognition and translation that involves concatenating existing data to improve model performance across multiple datasets and languages.

Contribution

The study introduces a minimal effort data augmentation technique using concatenation, which enhances speech recognition and translation models without requiring speaker information.

Findings

Significant WER reductions on LibriSpeech-960h test sets.

Up to 0.9 WER improvement on CoVoST-2 for multiple languages.

Method effective across speech recognition and translation tasks.

Abstract

Data augmentation is a technique to generate new training data based on existing data. We evaluate the simple and cost-effective method of concatenating the original data examples to build new training instances. Continued training with such augmented data is able to improve off-the-shelf Transformer and Conformer models that were optimized on the original data only. We demonstrate considerable improvements on the LibriSpeech-960h test sets (WER 2.83 and 6.87 for test-clean and test-other), which carry over to models combined with shallow fusion (WER 2.55 and 6.27). Our method of continued training also leads to improvements of up to 0.9 WER on the ASR part of CoVoST-2 for four non English languages, and we observe that the gains are highly dependent on the size of the original training data. We compare different concatenation strategies and found that our method does not need speaker information to achieve its improvements. Finally, we demonstrate on two datasets that our methods also works for speech translation tasks.