International Research Journal of Education and Technology

The only means of communication for someone who cannot hear or speak is through sign language. The use of sign language by people with physical disabilities is a great help in communicating their ideas and feelings. With the help of computer vision and neural networks we can detect the signs and give the respective text output. The project introduces an application of computer vision for American Hand sign detection. Research was carried out on a number of algorithms that could best differentiate a hand gesture. It was found our Long ShortTerm Memory. We use the Media-pipe Hand landmark detection algorithm to detect the hand landmarks plot the hand coordinates. Later the hand coordinates are mapped into NumPy array (Npy)format and feed to Long Short Term Memory model. The model predicts in real time and is able to convert text to speech. Keywords: Computer vision, LSTM, Hand gestures, Numpy array format