Abstract

Speech is an important mode of communication for humans with them as well as with machines in real time applications. In many interactive applications machines or systems has to process and recognize the speech signal as a source of data and it has to respond accordingly. This speech recognition will be handled by Automatic Speech Recognition system. These ASR will recognize the fluent speech, but if it is a disfluent speech the ASR requires more time or it may not be able to recognize the disfluent speech. Disfluency, is a speech disorder includes blockages, prolongations, repetitions, apraxia and interjections, also known as stuttered voice or stammering voice. In order to recognize the stuttered voices different features are extracted using MFCC,LPC,LPCC,DTW etc techniques and it will classified using ANN,HMM,KNN,SVM techniques.