With the advent of image processing and computer vision techniques, there has been huge research going on in the field of biometric classification and identification. On combining these techniques with machine learning, not only efficient but also smart applications can be developed. Biometric refers to measurement of various human body parts such as hands, fingers, face, etc. to distinctive features such as fingerprints, iris scan, etc. A sub branch of this field is hand gesture recognition which deals with classification of hand movements and signs. This paper shows how to classify those hand gestures using Computer Vision and Machine Learning algorithms and convert them into text and voice to facilitate communication between the hearing deficient and normal people.