A Study on Effectiveness of Deep Neural Networks for Speech Signal Enhancement in Comparison with Wiener Filtering Technique

This chapter intends to provide the optimum method between Wiener filtering and neural network for speech signal enhancement. A is highly susceptible various noises. Many denoising methods include removal of high-frequency components from original signal, but this leads parts signal. Thus, quality reduces which undesirable. Our main objective denoise while we enhance its quality. Two methods, namely, fully connected convolutional are compared with method, most suitable technique will be suggested. To compare output quality, compute signal-to-noise ratio (SNR) peak (PSNR). An advanced version MATLAB toolboxes such as Deep Learning toolbox, Audio Signal Processing etc. utilized