WACO: Word-Aligned Contrastive Learning for Speech Translation
This addresses the challenge of low-resource speech translation for languages with limited parallel data, though it appears incremental as it builds on contrastive learning techniques.
The paper tackles the problem of poor performance in end-to-end speech translation when only extremely small speech-text parallel data is available, and demonstrates that their Word-Aligned Contrastive Learning (WACO) method outperforms the best baseline by over 9 BLEU points with just 1-hour of training data.
End-to-end Speech Translation (E2E ST) aims to directly translate source speech into target text. Existing ST methods perform poorly when only extremely small speech-text data are available for training. We observe that an ST model's performance closely correlates with its embedding similarity between speech and source transcript. In this paper, we propose Word-Aligned COntrastive learning (WACO), a simple and effective method for extremely low-resource speech-to-text translation. Our key idea is bridging word-level representations for both speech and text modalities via contrastive learning. We evaluate WACO and other methods on the MuST-C dataset, a widely used ST benchmark, and on a low-resource direction Maltese-English from IWSLT 2023. Our experiments demonstrate that WACO outperforms the best baseline by 9+ BLEU points with only 1-hour parallel ST data. Code is available at https://github.com/owaski/WACO.