• Manthan Rane
• Triveni Gavathe
• Damini Tawde

- This project presents a real-time system that converts American Sign Language (ASL) into both text and speech using Convolutional Neural Networks (CNN). The objective is to bridge the communication gap between deaf and dumb individuals and those unfamiliar with sign language. Hand gestures are captured using a webcam, and the MediaPipe library is utilized to detect hand landmarks irrespective of background or lighting conditions. These landmarks are drawn on a plain background and processed using OpenCV for image pre-processing. A CNN model is trained on the processed data to classify hand gestures, initially grouping similar letters to improve accuracy. Further classification is done using mathematical operations on the hand landmarks to identify specific characters. Finally, the recognized text is converted into speech using the pyttsx3 library. The system achieved up to 99% accuracy under optimal conditions and maintains high accuracy in varying environments, offering a practical and accessible communication tool

Sign Language Recognition, Indian Sign Language (ISL), Convolutional Neural Network (CNN), Hand Gesture Recognition, Speech Conversion

"SIGN LANGUAGE TO TEXT AND SPEECH CONVERSION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.k375-k378, April-2025, Available :http://www.jetir.org/papers/JETIR2504A46.pdf

"SIGN LANGUAGE TO TEXT AND SPEECH CONVERSION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppk375-k378, April-2025, Available at : http://www.jetir.org/papers/JETIR2504A46.pdf