A Translation System That Converts English Text to American Sign Language Enhanced with Deep Learning Modules
Lalitha Natraj1, Sujala D. Shetty2
1Lalitha Natraj*, Birla Institute of Technology and Science – Pilani, Dubai, U.A.E.
2Sujala D. Shetty, Birla Institute of Technology and Science – Pilani, Dubai, U.A.E.
Manuscript received on September 16, 2019. | Revised Manuscript received on 24 September, 2019. | Manuscript published on October 10, 2019. | PP: 5378-5383 | Volume-8 Issue-12, October 2019. | Retrieval Number: L37811081219/2019©BEIESP | DOI: 10.35940/ijitee.L3781.1081219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: A recent surge in interest to create translation systems inclusive of sign languages is engendered by not only the rapid development of various approaches in the field of machine translation, but also the increased awareness of the struggles of the deaf community to comprehend written English. This paper describes the working of SILANT (SIgn LANguage Translator), a machine translation system that converts English to American Sign Language (ASL) using the principles of Natural Language Processing (NLP) and Deep Learning. The translation of English text is based on transformational rules which generates an intermediate representation which in turn spawns appropriate ASL animations. Although this kind of rule-based translation is notorious for being an accurate yet narrow approach, in this system, we broaden the scope of the translation using a synonym network and paraphrasing module which implements deep learning algorithms. In doing so, we are able to achieve both the accuracy of a rule-based approach and the scale of a deep learning one.
Keywords: Artificial Intelligence, Knowledge Based Systems, Natural Language Processing, Neural Networks, Sign Language.
Scope of the Article: Artificial Intelligence