DSP based improved Speech Recognition system

Processing of large amount of data is an important aspect of Speech Recognition (SR). However, to formulate speech recognition system in small devices is not simple. This paper suggests Digital Signal Processor (DSP) based speech recognition system with improved performance in terms of recognition accuracies and computational cost. The comprehensive survey of various approaches of feature extraction, like Mel Filter Banks with Mel Frequency Cepstrum Coefficients (MFCC) and Cochlear Filter Banks (CFB) with Zero-crossings is given. Amongst various feature classification techniques, the suitability of the Support Vector Machine (SVM) classifier for the proposed system is significant.