Comparative Analysis of Automatic Sign Language Generation Systems
Rakesh Kumar, Vishal Goyal, Lalit Goyal · Journal of scientific research · 2021
More than 466 million people suffer from hearing disabilities comprising of 34 million children and 432 million adults worldwide.These people face so many problems at public places due to their hearing disability for the communication of their messages.The development of automatic sign language generation systems, which are very rarely available, is the need of the hour to solve their communication problem.Some researchers have tried to develop such systems for various public places using a smaller corpus rendering their works limited to a specific domain.In this paper, we tried to perform a study of available research studies from 2001 to 2020 for sign language generation systems for various public domains using different machine translation approaches.Seventy-two research papers on sign language generation were retrieved from our search.Afterward, these papers were shortlisted to 44 based on their titles, then 30 research papers were removed based on abstracts and conclusions, and finally, 14 research papers were selected based on full text.These 14 research papers were comparatively examined based on six comparison parameters.It is evident from our study that the lack of proper grammar rules of sign language and the non-availability of large bilingual corpora are the main hurdles in developing sign language generation systems for public places.A hybrid approach based announcement system prototype is developed for deaf people.The proposed system prototype produced 82% accuracy on translation of announcements into sign language.