Analysis and Implementation of Speech Recognition System using ARM7 Processor

Abstract

This paper introduces implementation and analysis of speech recognition system. Speech Recognition is the process of automatically recognizing a certain word spoken by a particular speaker based on individual information included in speech waves. This paper presents one of the techniques to extract the feature set from a speech signal, which can be used in speech recognition systems and an analysis study has been performed. A wide range of possibilities exist for parametrically representing the speech signal for the speaker recognition task, such as Linear Prediction Coding (LPC), Mel-Frequency Cepstrum Coefficients (MFCC),and others. Studies and experiments show that MFCC provides better results than LPC. Here vector quantization is used to increase speech recognition accuracy. Experiments shows that as the no. of MFCC coefficients increases get better accuracy, code book size also affects accuracy. The MFCC and VQ algorithm, for speech recognition have been implemented in MATLAB 7.7(R2008b) version on Windows7 platform. The control circuitry has been implemented in Keil µVision3; the supporting hardware setup is being implemented.

Keywords: Speech Recognition; MFCC; Vector Quantization; LPC