A Survey Report on Speech Recognition System

Speech Recognition is the process of converting an acoustic waveform into text containing the similar information conveyed by speaker. This paper present a report on a Automatic Speech Recognition System (ASR) for different language under different accent. The paper describe the methods used and comparative study of the performance of every system so far developed. The study shows that Hidden Markov Model(HMM) as classifier and Mel Frequency Cepstral Coefficients(MFCC) as speech features are the most common technique used. And Moreover ASR implemented by using Hidden Markov Tool kit(HTK) are more efficient then the other systems implemented by using other tools