Tri-Control

This paper introduces a comprehensive multimodal system designed for accessible computer control, incorporating eye movement tracking, gesture recognition based on sign language, and voice commands. The system aims to provide a hands-free, user-friendly interface that significantly enhances digital accessibility for users with disabilities, such as motor impairments or vision-related limitations, as well as users seeking efficient multitasking solutions. Implemented using Python and associated libraries (OpenCV, TensorFlow, SpeechRecognition), the platform integrates real-time computer vision, machine learning, and speech recognition technologies. The result is a unified interaction model that empowers users to perform cursor movements, clicks, and various computing tasks with precision and convenience. Extensive testing demonstrated accuracy levels above 90% across different modalities. The system promotes digital inclusion and provides a foundation for scalable assistive technology.