• Medhini Prabhakar
• Prasad Hundekar
• Sai Deepthi B P
• Shivam Tiwari
• Vinutha M S

This system presents a novel approach for translation of sign action analysis, recognition and generating a text description in English language and then conversion of the generated text to speech. In training set there were 26 Indian Sign Language Alphabet image samples used whereas testing captures hand gestures from the live feed and predicts the class label based on several trained models like CNN (Convolutional Neural Networks), FRCNN(Faster-Convolutional Neural Networks), YOLO(You Only Look Once) and Media Pipe. Finally, the text description will be generated in English language and converted to speech. The average computation timeis bit more than expected due unavailability of high GPU hardware but has acceptable recognition rate in case of FRCNN model. When it comes to CNN model, the recognition of hand gestures is fast enough for real world applications but there is a compromise in accuracy of identification. YOLO model recognizes the sign language with a good accuracy but is not satisfactory in case of speed as it is taking more time when live feed of hand gestures is captured and converted. Though YOLO model works inefficiently for real time conversion, it performs greatly when already captured hand gestures are fed as input to the model. Media Pipe model of sign language conversion checks all the requirements of our project which include great accuracy as well as real time conversion of hand gestures to text to speech in real time without any delay.

Convolutional Neural Networks(CNNs), FRCNN(Faster-CNN), YOLO(You Only Look Once) , Media Pipe

"Sign Language Conversion To Text And Speech", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 7, page no.d621-d626, July-2022, Available :http://www.jetir.org/papers/JETIR2207380.pdf

"Sign Language Conversion To Text And Speech", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 7, page no. ppd621-d626, July-2022, Available at : http://www.jetir.org/papers/JETIR2207380.pdf