Projects per year
Abstract
This paper describes Edinburgh’s submissions to the IWSLT2021 multilingual speech translation (ST) task. We aim at improving multilingual translation and zero-shot performance in the constrained setting (without using any extra training data) through methods that encourage transfer learning and larger capacity modeling with advanced neural components. We build our end-to-end multilingual ST model based on Transformer, integrating techniques including adaptive speech feature selection, language-specific modeling, multi-task learning, deep and big Transformer, sparsified linear attention and root mean square layer normalization. We adopt data augmentation using machine translation models for ST which converts the zero-shot problem into a zero-resource one. Experimental results show that these methods deliver substantial improvements, surpassing the official baseline by > 15 average BLEU and outperforming our cascading system by > 2 average BLEU. Our final submission achieves competitive performance (runner up).
Original language | English |
---|---|
Title of host publication | Proceedings of the 18th International Conference on Spoken Language Translation (IWSLT 2021) |
Place of Publication | Bangkok, Thailand (online) |
Publisher | Association for Computational Linguistics |
Pages | 160-168 |
Number of pages | 9 |
ISBN (Electronic) | 978-1-954085-74-9 |
DOIs | |
Publication status | Published - 5 Aug 2021 |
Event | 18th International Conference on Spoken Language Translation - Online Duration: 5 Aug 2021 → 6 Aug 2021 https://iwslt.org/2021/ |
Conference
Conference | 18th International Conference on Spoken Language Translation |
---|---|
Abbreviated title | IWSLT 2021 |
Period | 5/08/21 → 6/08/21 |
Internet address |
Fingerprint
Dive into the research topics of 'Edinburgh's End-to-End Multilingual Speech Translation System for IWSLT 2021'. Together they form a unique fingerprint.Projects
- 1 Finished