Review of Single Channel Speech Enhancement Methods in Spectral Domain
Anuprita P. Pawar, Kirtimalini B. Choudhari, Madhuri A. Joshi · 2012
Main objective of speech enhancement is to improve perceptual aspects of speech such as overall quality, intelligibility and degree of listener fatigue. The Spectral Subtraction method is historically one of the first algorithms proposed by S.F Boll in 1979. In this method, an estimated speech signal is simply obtained by subtracting a preestimated noise spectrum from an observed one. It is a single channel enhancement method for the enhancement of speech degraded by additive background noise. Additive background noise affects our conversation in noisy environment like in streets or in car, when sending message from the cockpit of an airplane to the ground or cabin. With the passage of Spectral Subtraction has undergone many modifications. In this paper we present the review of various Spectral Subtraction methods.