Analysis of Different Discrete Wavelet Transform Basis Functions in Speech Signal Compression
Student(IV), Dept of ECE, LIET, Vizianagaram, Andhra Pradesh, India, Sai Lakshmi Bhamidipati, Sai Sudha Mindagudla, Harsha Vardhan Devalla, Hima Sagar Goodi, Hemanth Nag · IOSR Journal of VLSI and Signal processing · 2014
In this paper we attempt to evaluate the challenge of compression of speech signals.Compression of speech signals have pre-dominantly occupied a considerable position in the present era of multimedia.Speech compression is one area of digital signal processing that focuses on reducing the bit rate of the speech signal for transmission or storage without significant loss of quality.In recent years a new technique called wavelet transform has been proposed for signal analysis.It has been successfully used in image compression application.So far, less attention has been paid to the research in the speech compression using wavelet.Here we assess the compression of speech signal using different discrete wavelet transform basis functions.There are different wavelet basis families like haar wavelet, daubechies wavelet, biorthagonal spline wavelet, coiflet wavelet, meyer wavelet, reverse biorthogonal wavelet, Shannon wavelet, symlet wavelet.The auditory masking method and psycho acoustic methods are used to compress the speech.At first the speech is divided into number of frames and upon each frame wavelet transformation is used to minimize number of bits required to represent frame while keeping any distortion inaudible.MATLAB code is implemented to perform the compression.This paper performs evaluation of different wavelet families on the basis of compression scores each contributing itself in the compression of speech.