Abstract: Creating a desktop application that uses a computer webcam to capture a person signing gestures for Indian Sign Language (ISL) and translate it into corresponding text and speech in real time. The translated sign language gesture will be acquired in text which is further converted into audio. In this manner we are implementing a finger spelling language translator. To enable the detection gestures, we are making use of a Convolution neural network (CNN). A CNN is highly efficient in tackling computer vision problems and is capable of detecting the desired features with high degree of accuracy upon sufficient training. This project is about converting the hand gesture of sign language to voice or text using Machine Learning Techniques and vice versa. In this we are going to capture a real time translation of indian sign language using single and double hand gestures and recognize the words and convert it into text and then to speech. If the person gives speech as input it is first converted to text and then it displays the suitable sign as output and vice versa.

Keywords: Indian Sign Language, Hand Gesture Recognition, Convolution Neural Network, K- means algorithm, Open CV.

| DOI: 10.17148/IARJSET.2021.8412