A Proportional Study on Feature Extraction Method in Automatic Speech Recognition System

Automatic speech recognition (ASR) has been the focus of many researchers for several years. In speech recognition system is for a computer be able to "hear,” understand," and "act upon" spoken information. The speaker recognition system viewed as working in a Analysis , Feature extraction , Modeling , Testing/Matching techniques .speech processing is to convey information about words, speaker identity, tone of voice, expression, style of speech, emotion and the state of health of the speaker. Speech features can be extracted by the parameter as Linear Predictive Codes (LPC), Perceptual Linear Prediction (PLP), Mel Frequency Cepstral Coefficients (MFCC), PLP-Relative Spectra .Neural network training like Vector Quantization (VQ), Dynamic Time Warping (DTW), Support Vector Machine (SVM), and Hidden Markov Model (HMM),Gaussian Mixture Module (GMM) can be used for classification and recognition technique, We have to illustrate some of most important method in neural network like LPC, MFCC, and PLP.