Self-Training for End-to-End Speech Translation
Preprint
Abstract
One of the main challenges for end-to-end speech translation is data scarcity. We leverage pseudo-labels generated from unlabeled audio by a cascade and an end-to-end speech translation model. This provides 8.3 and 5.7 BLEU gains over a strong semi-supervised baseline on the MuST-C English-French and English-German datasets, reaching state-of-the art performance. The effect of the quality of the pseudo-labels is investigated. Our approach is shown to be more effective than simply pre-training the encoder on the speech recognition task. Finally, we demonstrate the effectiveness of self-training by directly generating pseudo-labels with an end-to-end model instead of a cascade model.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Sequence-to-Sequence Speech Recognition with Time-Depth Separable Convolutions2019
- Wav2Letter++: A Fast Open-source Speech Recognition System2019
- Accelerating Neural Transformer via an Average Attention Network2018
- SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing2018
- Sequence-to-Sequence Models Can Directly Translate Foreign Speech2017
- Probability of error of some adaptive pattern-recognition machinesIEEE Transactions on Information Theory, 1965