Hardware Implementation of Speech Recognition Using MFCC and Euclidean Distance

ABSTRACT:This paper suggests Digital Signal processor (DSP) based speech recognition system with improved performance in terms of recognition accuracies and computational cost. The comprehensive surrey of various approaches of feature extraction like Mel filter banks with Mel Frequency Cepstrum Coefficients (MFCC). This paper describes an approach of isolated speech recognition by Digital Signal Processor TMS320C6713 using Mel scale Frequency Cepstral Coefficients and Euclidean distance. Several features are extracted from speech signal of spoken words. An experiments database of total five speakers, speaking 5-10 words each is collected under acoustically controlled room is taken. MFCC are extracted from speech signal of spoken words. To compare inter speaking differences Euclidean distance is used KEYWORDS:Speech Recognition, Feature Extraction MFCC, Pattern Recognition, Euclidean distance. I.INTRODUCTION Speech Recognition is the process of automatically recognizing the spoken words of person based on information in speech signal. Each spoken word is created using the phonetic combination of a set of vowel, semivowel and consonant speech sound units. The popular spectral based parameter used in recognition approach is the Mel Freqency Cepstral Coefficients called MFCC, MFCC's are coefficients, which represents audio, based on perception of human auditory systems. The basic difference between the operation of FFT/DCT and the MFCC is that in the MFCC, the frequency bands are positioned logarithmically (on the mel scale) which approximates the human auditory system's response more closely than the linearly spaced frequency bands of FFT or DCT. II.LITERATURE SURVEY The system consists of microphone through which input in the form of speech signal is applied. The data acquisition system of speech processor acquires the output from the microphone and then itdetects the exact word spoken If such a system is installed in a motor car, then by using several start, stop,forward,backwardetc.; we can drive a carwithout even out hands. It is widely applied to some special fields such as resource poor embedded system for its simple and effective algorithm.DTW-based application of portable value added tax calculator with speaker dependent connected word and isolated word speech recognition abilities built in. [2]Automatic speech recognition is an interesting task but it requires a lot of effort. With technical development, speech recognition system achieved excellent results, still it possess some major limitations. Especially, recognition system with hidden markov models (HMM) as major elements are suitable for many applications, but do not suffer from major restrictions that make speech recognition system unsuitable for real time applications An isolated word, speaker dependent speech recognition system capable of recognizing spoken words at sufficiently high accuracy. The system has been tested and verified on MATLAB as well as TMS320C6713 DSK wit