Machine Learning techniques gives computers the capability to learn using sample inputs and their outputs which creates a model to test against test cases instead of being explicitly programmed. Visual speech recognition is a process of conversion of speech to text in the absence of audio where the lip features of the person are extracted to track the pattern formed. This paper also contains the overview of different Machine Learning algorithms and image processing procedures to effectively

more »

... to effectively extract and track the lip movements. Nowadays image processing procedure is turning into a key technique for extracting the key features and considering various other environmental features to enhance the output. This paper predominantly centers prediction of text based on the lip movements. This paper mainly focuses on reviewing various algorithms used for VSR. There are so many classification techniques such as LSTMs, CNNs, Decision Tree and Neural networks.