Abstract
Recently, end-to-end speech translation (ST) has gained significant attention as it avoids error propagation. However, the approach suffers from data scarcity. It heavily depends on direct ST data and is less efficient in making use of speech transcription and text translation data, which is often more easily available. In the related field of multilingual text translation, several techniques have been proposed for zero-shot translation. A main idea is to increase the similarity of semantically similar sentences in different languages. We investigate whether these ideas can be applied to speech translation, by building ST models trained on speech transcription and text translation data. We investigate the effects of data augmentation and auxiliary loss function. The techniques were successfully applied to few-shot ST using limited ST data, with improvements of up to +12.9 BLEU points compared to direct end-to-end ST and +3.1 BLEU points compared to ST models fine-tuned from ASR model.
Original language | English |
---|---|
Title of host publication | ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
Publisher | IEEE |
Pages | 6222-6226 |
Number of pages | 5 |
ISBN (Print) | 9781665405409 |
DOIs | |
Publication status | Published - 2022 |
Event | 47th IEEE International Conference on Acoustics, Speech and Signal Processing - Online, Singapore, Singapore Duration: 22 May 2022 → 27 May 2022 Conference number: 47 https://2022.ieeeicassp.org/ |
Publication series
Series | International Conference on Acoustics Speech and Signal Processing Proceedings |
---|---|
ISSN | 1520-6149 |
Conference
Conference | 47th IEEE International Conference on Acoustics, Speech and Signal Processing |
---|---|
Abbreviated title | ICASSP 2022 |
Country/Territory | Singapore |
City | Singapore |
Period | 22/05/22 → 27/05/22 |
Internet address |
Keywords
- speech translation
- zero-shot
- few-shot
- machine translation
- multi-task