Sign language is a common language that deaf people around the world use to communicate with others. However, normal people are generally not familiar with sign language (SL) and they do not need to learn their language to communicate with them in everyday life. Several technologies offer possibilities for overcoming these barriers to assisting deaf people and facilitating their active lives, including natural language processing (NLP), text understanding, machine translation, and sign language simulation. In this paper, we mainly focus on the problem faced by the deaf community in Saudi Arabia as an important member of the society that needs assistance in communicating with others, especially in the field of work as a driver. Therefore, this community needs a system that facilitates the mechanism of communication with the users using NLP that allows translating Arabic Sign Language (ArSL) into voice and vice versa. Thus, this paper aims to purplish our created dataset dictionary and ArSL corpus videos that were done in our previous work. Furthermore, we illustrate our corpus, data determination (deaf driver terminologies), dataset creation and processing in order to implement the proposed future system. Therefore, the evaluation of the dataset will be presented and simulated using two methods. First, using the evaluation of four expert signers, where the result was 10.23% WER. The second method, using Cohen’s Kappa in order to evaluate the corpus of ArSL videos that was made by three signers from different regions of Saudi Arabia. We found that the agreement between signer 2 and signer 3 is 61%, which is a good agreement. In our future direction, we will use the ArSL video corpus of signer 2 and signer 3 to implement ML techniques for our deaf driver system.

Arabic Sign Language (ArSL) is a formal language used by Deaf and Hard of Hearing communities. It helps them understand Arabic spoken by hearing people, particularly in formal speech, such as news or sermons. The production of sign language systems can support communication by translating spoken languages into ArSL videos. However, current systems often fail to meet user needs due to poor translation of grammatical structures and limited ability to capture the textual meaning of speech. This research focused on a neural machine translation (NMT) approach to convert spoken Arabic texts into Arabic Sign Language (ArSL) using a transformer-based model to overcome these difficulties. The ArSL video of the sign is produced as a series of 2D skeletal poses. The proposed methodology is based on an encoder-decoder architecture: texts are represented by a transformer model, and a posture decoder generates gestures. The model is evaluated using quantitative criteria, such as mean square error (MSE) and the dynamic time warping (DTW) measure, as well as qualitative evaluations by sign language experts and users from the Deaf and Hard of Hearing community. The study uses Friday sermon data as a reference to build the translation model and applies optimization techniques such as temporal modulation, noise addition, and resizing to enhance performance. The results show that the enhanced model improves the quality of gesture representation, reducing DTW by 43.4% relative to the baseline in the non-continuous ArSL test. On the other hand, in the Continuous ArSL test using Short Sermon, the DTW is reduced by around 2.25%. In addition, the enhanced model received high user acceptance, with 75.2% satisfaction. Moreover, unified qualitative evaluation criteria for sign language production are proposed, including clarity, accuracy, consistency, and naturalness, which could serve as a basis for future evaluations in this field. These criteria are used to evaluate the generated ArSL and measure user satisfaction.