Speech Enhancement for Quality and Intelligibility Enhancement under Different Noise Conditions

Abstract

Speech intelligibility is an important measure of the effectiveness or adequacy of a communication system or to communicate in a noisy environment. In many daily life situations it is important to understand what is being said, for example over a loudspeaker system, and to be able to react to acoustic signals of different kinds. Inherent noise presence in message production process has huge impact. Both production and interpretation noise processes have a fixed signal-to-noise ratio. A simple and effective model of human communication is proposed in this work which has ability to enhance the intelligent speech in noisy environment. Simulation results yields better performance over traditional state-of-methods.

Keywords: Speech Communication, Noise Environment, Intelligibility Enhancement

1. INTRODUCTION

Speech output – whether from mobile phones, public address systems or simply domestic audio devices – is widely used. In many listening contexts the intelligibility of the intended message might be compromised by environmental noise or channel

 distortion. Problems can be minimized by increasing output intensity or repeating the message, but these approaches are not ideal for either the listener or the output device (e.g. power consumption, failure). A better approach is to seek ways to modify the speech signal to increase intelligibility in noise.