Voice activity Detection Using Wavelet Transform

In this paper, we look at 2 different methods for Voice Activity Detection (VAD) based on the Wavelet Transform. The algorithms utilize the Wavelet Transform’s flexibility in the time-frequency resolution to compute robust parameters for VAD decision. We examine the performance of the algorithms in clean and various noisy environments with varying levels of noise main.m : To run the two algorithms on all the audio signals and generate the comparison plots.
single_file.m : To observe the algorithm in action on a single audio file
babble.mat factory1.mat white.mat leopard.mat : Noise signals