Performance Analysis of Speech Performance Analysis of Speech Recognition using Advanced Algorithm

In this paper, the authors have worked on a speech processing technique wherein they took audio speech as input signal from user and by recognition process, they perceived recorded word. Specifically, the authors concentrated on the amplitude and zero cross of speech signal, and two of the most important and useful terms, namely MFCC and Mahalanobis distance, for the purpose of getting an efficient outcome. The authors evaluated the proposed method with some quality matrices of different speech signals and based on the results they concluded that the proposed method provides accurate process of speech recognition against various models for comparing speech processing.