<p>This article aims to present a novel Arabic sign language recognition (SLR) strategy using sensory glove and machine learning. The article focuses on hand gesture recognition through the development of a glove-computer system designed for real-time hand posture detection and gesture-to-text translation. Gesture recognition plays a crucial role in enhancing interactions between humans and machines, making technology more intuitive and efficient. This technology has potential applications in various fields such as smart homes, gaming, automotive systems, and virtual reality. The primary goal of this article is then to create a supportive communication environment for individuals with speaking difficulties. The article began with the development of a sensory glove equipped with sensors to detect hand orientation and finger flexing, with data processed and transmitted wirelessly to a computer for machine learning prediction. A dynamic dataset, which included signs for letters and movement-based signs for words, was created and used to build two machine learning models: Support Vector Machine (SVM) model with feature extraction (SVM-FE model) and Long Short-Term Memory (LSTM) model. The proposed deep learning LSTM model demonstrated superior performance with accuracy of 99.6%. Based on these findings, a real-time recognition application was developed using the LSTM model, effectively showcasing the system's practical applicability in real-world scenarios.</p>