Translation from Spoken Arabic Digits to Sign Language based on Deep Learning
Laith Abualigah, Mothanna Almahmood, Hazem Migdady, Raed Abu Zitar, Aseel Smerat · 2026
Deaf-and-dumb humans make up about 5% of the world&s;s population, and they need special care by providing alternative methods that help them communicate with the outside world. In contrast, the sense of hearing is the main element of human communication, which is indispensable. From the standpoint of introducing helpful applications that help the deaf-and-dumb population, the idea of this research aimed to use deep learning techniques to create a model based on the principle of converting Arabic spoken digits to sign language images through a study of two different datasets that were freely taken from open-source websites. The first one contains audio records of Arabic spoken digits that were used to train the on-dimensional CNN model to generate a text translation of any Arabic spoken digit record. The second one contains sign language images of Arabic digits, where used to build an IF-THEN rules system that can generate the sign language image as a translation of given Arabic digit text. The whole idea was conducted by using both systems in one prediction model that can generate the sign language image of any given spoken Arabic digits’ record, where it had accurate results with 86.85% accuracy value and 0.5039 loss value. The goal of this research is to add a new technology based on deep learning in order to help this group of people with a simple idea that opens the researchers’ minds to produce a model of all Arabic spoken speech, which in turn can be a complete technology that helps deaf-and-dumb humans to communicate with the outside world easily.