Automation: A Tool for the Evolution of a Portuguese-Libras Automatic Translator with Collaborative Corpus

• Marcos Henrique Alves da Silva UFPB
• Tiago Maritan Ugulino de Araújo UFPB
• Rostand Edson Oliveira Costa UFPB
• Samuel de Moura Moreira UFPB
• Lucas Moreira e Silva Alves UFPB

DOI: https://doi.org/10.5753/semish.2021.15818

Abstract

Deaf people communicate naturally using visuospatial languages, called Sign Languages (SL). Although Sign Languages are recognized in many countries as a second official language, content is not always generated in an accessible way for this community. In order to minimize these impacts, research involving automatic translation using Deep Learning is gaining more and more space in the literature. However, one of the main barriers to using this strategy is the need for a bilingual corpus, which is not always an easy task to find or produce. In this context, this paper presents a training automation tool for a Portuguese-Libras automatic translator. The solution communicates with a collaborative system of translations of Portuguese-Libras sentences, WikiLibras, and feeds the corpus of an automatic translator for Libras, VLibras, with the goal of conducting training with an increasingly robust data set each iteration.

Keywords: Machine Translation, Corpus, Libras, Automation

References

Araujo, T. (2012). Uma solução para geração automática de trilhas em Língua Brasileira de Sinais em conteúdos multimídia. PhD thesis

Bisong, E. (2019). Google Colaboratory, pages 59–64. Apress, Berkeley, CA.

Krishna, S., Jindal, A. R., R, M., K, R., and Jayagopi, D. (2020). Virtual indian sign language interpreter. In Proceedings of the 2020 4th International Conference on Vision, Image and Signal Processing, ICVISP 2020, New York, NY, USA. Association for Computing Machinery.

Lima, M. A. C. B. (2015). Tradução automática com adequação sintático-semântica para LIBRAS. PhD thesis.

Liu, D., Ma, N., Yang, F., and Yang, X. (2019). A survey of low resource neural machine translation. In 2019 4th International Conference on Mechanical, Control and Computer Engineering (ICMCCE), pages 39–393.

Ott, M., Edunov, S., Baevski, A., Fan, A., Gross, S., Ng, N., Grangier, D., and Auli, M. (2019). fairseq: A fast, extensible toolkit for sequence modeling. In Proceedings of NAACL-HLT 2019: Demonstrations.

Papineni, K., Roukos, S., Ward, T., and Zhu, W.-J. (2002). Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylva- nia, USA. Association for Computational Linguistics

Rosa Zucolotto, M.P., Ruiz, L.R., and Pinheiro, N.F. (2019).Reflexões sobre linguagem, sociedade e surdez. Revista Uniabeu, 12(30):134–147.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need.

Veríssimo, V., Silva, C., Hanael, V., Moraes, C., Costa, R., Maritan, T., Aschoff, M., and Gaudêncio, T. (2019). A study on the use of sequence-to-sequence neural networks for automatic translation of brazilian portuguese to libras. In Proceedings of the 25th Brazillian Symposium on Multimedia and the Web, WebMedia ’19, page 101–108, New York, NY, USA. Association for Computing Machinery

Vichyaloetsiri, T., Wuttidittachotti, P., and (2017). Web service framework to translate text into sign language. In 2017 International Conference on Computer, Information and Telecommunication Systems (CITS), pages 180–184.

Wu, F., Fan, A., Baevski, A., Dauphin, Y. N., and Auli, M. (2019). Pay less attention with lightweight and dynamic convolutions. CoRR, abs/1901.10430.