This paper presents the development of an Arabic Sign Language ArSL translator application leveraging machine learning techniques. The motivation is to bridge the communication gap between the deaf community and others by providing an accessible and user-friendly solution for real-time sign language translation. The methodology involved collecting and preprocessing an ArSL image dataset training and evaluating multiple machine learning models including k-nearest neighbors support vector machines decision trees multi-layer perceptrons and convolutional neural networks. The convolutional neural network model integrated with Amazon Rekognition achieved the highest accuracy of 98% in recognizing ArSL gestures. The application was developed with the AWS architecture to enable real-time translation. The ArSL translator app aims to support sustainable community development by promoting inclusion education healthcare access and economic opportunities for the deaf community in alignment with multiple UN Sustainable Development Goals. © 2024 IEEE.