Multimodal Communication for Deaf and Dumb

Communication barriers often hinder effective interaction for individuals with hearing and speech impairments. To address this challenge, we propose a Multimodal Communication System that leverages advanced technologies such as Convolutional Neural Networks (CNNs), sign recognition, and real-time communication capabilities. The system employs 8 classes of sign language gestures, each containing 300 curated images, as a training dataset for the CNN s. Through iterative training, the CNN s learn to interpret sign language gestures and voice commands, providing users with multiple avenues for communication. In instances where a sign is not detected accurately, the system prompts users for corrective action, ensuring smoother interactions. The integration of an emergency notification system further enhances safety and responsiveness, fostering inclusivity and empowerment for individuals with disabilities.