Real-time sign language recognition has become a crucial area of research, enabling more inclusive interactions across various digital platforms. This study compares YOLOv6 and YOLOv8, focusing on their respective architectures and performance metrics when applied to ASL recognition. Experiments are conducted on a robust dataset of hand gestures, with results analyzed for accuracy, inference speed, and resource efficiency. Our findings demonstrate that YOLOv8 achieves a balance of high accuracy, speed, and efficiency, making it particularly suitable for mobile applications. The research underscores the advances in real-time object detection technology, particularly YOLO’s adaptability for gesture recognition tasks, and its potential in enhancing accessibility tools across diverse platforms.