Transfer Learning for British Sign Language Modelling

Boris Mocialov, Helen Hastie, Graham Turner


Abstract
Automatic speech recognition and spoken dialogue systems have made great advances through the use of deep machine learning methods. This is partly due to greater computing power but also through the large amount of data available in common languages, such as English. Conversely, research in minority languages, including sign languages, is hampered by the severe lack of data. This has led to work on transfer learning methods, whereby a model developed for one language is reused as the starting point for a model on a second language, which is less resourced. In this paper, we examine two transfer learning techniques of fine-tuning and layer substitution for language modelling of British Sign Language. Our results show improvement in perplexity when using transfer learning with standard stacked LSTM models, trained initially using a large corpus for standard English from the Penn Treebank corpus.
Anthology ID:
W18-3911
Volume:
Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2018)
Month:
August
Year:
2018
Address:
Santa Fe, New Mexico, USA
Editors:
Marcos Zampieri, Preslav Nakov, Nikola Ljubešić, Jörg Tiedemann, Shervin Malmasi, Ahmed Ali
Venue:
VarDial
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
101–110
Language:
URL:
https://aclanthology.org/W18-3911
DOI:
Bibkey:
Cite (ACL):
Boris Mocialov, Helen Hastie, and Graham Turner. 2018. Transfer Learning for British Sign Language Modelling. In Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2018), pages 101–110, Santa Fe, New Mexico, USA. Association for Computational Linguistics.
Cite (Informal):
Transfer Learning for British Sign Language Modelling (Mocialov et al., VarDial 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-1/W18-3911.pdf