SignSense: Indian Sign Language Interpreter in Your Preferred Language

Accurate sign language interpretation is essential for bridging communication gaps between the deaf and hard-ofhearing community and non-signers. This work presents a realtime Indian Sign Language (ISL) Interpreter Web Application, leveraging deep learning-based image recognition to enhance accessibility and inclusivity. The system utilizes a fine-tuned Convolutional Neural Network (CNN) for gesture recognition, achieving an average training accuracy of $\mathbf{9 6. 8 8 \%}$, a validation accuracy of 99.26 %, and testing accuracy of 98.89 %. The model is trained on a diverse ISL dataset with extensive data augmentation to improve generalization and robustness. Users can capture ISL gestures through their device camera, with the system providing real-time interpretation in their preferred spoken language. This user-centric approach ensures personalized communication experiences, fostering seamless interaction between signers and non-signers. By combining advanced image processing techniques with deep learning, this solution significantly improves communication accessibility and promotes social inclusion for individuals with hearing impairments.