limited Arabic Alphabet Symbols Based on Lips Movement

Abstract

In this research, the development of deep learning algorithms was relied upon and employed in the field of computer vision for the purpose of building symbols for the Arabic alphabet with a limited capacity to cover specific requirements.
The network (CNN) is adopted for the purpose of tracking the movement of the lips when pronouncing selected words in the Arabic language by showing the borders of the lips (Lips land marks) and calculating the distance between the lips to determine the state of the mouth (open / closed) when forming a code for the selected words from the Arabic language and according to work requirements Where a special binary code is generated by tracking the movement of the lips, since the closed mouth is zero and the open mouth is the one as a basis for building codes for those movements.