An Avatar Based Translation System from Arabic Speech to Arabic Sign Language for Deaf People

The advancement of Information and Communication Technology has impacted all aspects of the human life. It has changed the way we work, travel, conduct business, study and communicate. For the Deaf community, we try to use ICT to improve their quality of life by developing systems that can help them communicate better with the rest of the world and amongst themselves. In this paper we present to the reader our research on the development of an Avatar based translation system from Arabic Speech to the Arabic Sign language for the Deaf people. The paper starts with an introduction of the problem of the Deaf community in the Arabic speaking community and a brief review of similar and related research. The next section describes the design consideration of the proposed system. The system will be composed of a database of captured 3D motions of Arabic sign language. The sign language motion will be recorded using data gloves. A Graphical translation of the digitized sign language will be re-animated using standard. The emphasis is to produce the original gesture as close as possible. Common spoken words are directly translated into respective semantic using Sphinx-4 Speech Recognition Engine without first being translated into text. Using the same Sphinx-4 Engine, the semantic of the spoken Arabic language will then be translated into Arabic sign language. With minimum supervision, the automation of the translation for converting standard Arabic speech into virtual Arabic Sign Language will be useful to the hearing impaired Arabic speaking community. Finally we describe future enhancements and other research projects that can be derived from this research.