About the Project

About the Project

As well as all spoken languages, Sign Languages SL have a structured grammar and syntax. Despite that it was visual, multi-dimensional and mainly based on gestures, SL follows specifics grammatical rules. Consequently, the automatic generation of sign language should follow these rules. It’s for this reason that Mada works on the development of a new framework that aims to support researchers and developers to create new innovant tools for deaf. The objective is essentially to create tools that enhance the development of software using grammatically validated sentences.


SL processing is a very hard task. It needs to be based on linguistics rules. However, many avatar-based translation systems use a word by word translation. In fact, this is due to the lack of linguistic studies on ArSL. Consequently, the lack of SDK and programming tools that help developers to build their own applications.


The main objective of this project is to provide to researchers and computer engineers a platform that offers a set of tools and data useful for processing and analyzing ArSL. Building a labeled database is fundamental to make any innovation in relation to the ArSL. In the first step, Mada will build the first labeled and indexed ArSL corpus containing both video and Mocap data. This content is very important for linguistics and computer researchers to study ArSL and make linguistic and scientific research. It can also be used by engineers to build new tools based on Mada corpus. In order to build this database a Sign Language studio will be setup using a dedicated Motion Capture solution that combines handshape, body movements and facial expression tracking.

The second objective concerns the translation to ArSL, it aims to build a new tool that converts texts to ArSL. The purpose of this tool is to provide a web service to developers for free. The service is offered to any developer after submitting an online request.