The deaf and mute individuals are a very important part of our society. Due to lack of aware-ness of sign languages among the general population, these individuals are often isolated. An effective solution for breaking this barrier is needed. In this paper we have discussed the implementation of a sign language translator by using Convolutional Neural Network (CNN). A few preprocessing techniques such as Skin Masking and Canny Edge Detection is applied. Finally the model of Convolutional Neural Network is applied by taking into account its recent advances for better detection of hand movements and classification. It has an excellent performance in machine learning problems.