Text to Pakistan sign language automation
No Thumbnail Available
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
UMT Lahore
Abstract
In this thesis, a text to Pakistan Sign Language automation system has been proposed. The respective system takes an English input sentence and converts it into a Pakistan Sign Language sentence. By considering an effective and efficient use of storage, computation, and bandwidth, animation is an appropriate and sophisticated way to keep sign language content on the Internet. Out of all present sign language writing notation systems, HamNoSys is an efficient available notation that is not only able to convert signs into animations but also supports non-manual features such as body, head, and shoulder movements with the help of the SiGML markup language. HamNoSys basically symbolizes the gestures of signs present.
The proposed system is not only capable of converting manually input text into its respective PSL version, but it also facilitates a deaf user to input any type of content from any web page. Our system is a HamNoSys-based system that is capable of generating animations of corresponding signs in PSL. As a benchmark, we took videos from PSL.org as standard gestures for our dictionary. Among all text to sign language conversion approaches, we followed a transfer-based translation approach, as we are not only translating input text into PSL but also validating and verifying the grammar of the converted sentence as well.