• JATIN JAGANI
• YOGITA PARIKH
• Vishal Savaliya
• Gaudani Mohit

This paper presents approach to convert "Sign Language into speech using Artificial Intelligence in real time" for deaf and dumb people. Deaf and Dumb People feels difficulty to communicate with the ordinary people. The sign language of deaf and dumb is quite difficult to learn and it is not possible for everybody to learn that language. This approach helps to provide communication between people with speech impairment and normal people, thereby reducing the communication gap between them. This system capable of recognizing 26gestures of letter from the Indian Sign Language by using Artificial Intelligence. The proposed system having Real time video acquisition, frame extraction and Pre-processing, Hand gesture segmentation, Sign recognition using AI, Text and audio output of letter. By using Webcam, capture the real time video of hand gestures for Indian sign language and extract the video frames from it. Brightness and orientation of frame is derived and if its value is not proper, then frame is rejected and it is acquired again. Sign gesture is segmented by using K-means clustering algorithm and morphological image processing. Sometimes there is difficult to segment the hand gestures due to different background conditions. A convolutional Neural Network recognize the letter of ISL with the 97% recognition rate. It is trained by using different segmented hand gesture of letters in different background conditions. Recognized gesture is used to convert signinto text and audio output for users.

Sign language converter, Artificial Intelligence, Text to speech conversion, Indian Sign Language.

"Real time Indian sign language system using Artificial Intelligence ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.586-593, April-2019, Available :http://www.jetir.org/papers/JETIR1904N80.pdf

"Real time Indian sign language system using Artificial Intelligence ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp586-593, April-2019, Available at : http://www.jetir.org/papers/JETIR1904N80.pdf