Speech enhancement in a Bayesian framework

WC present an approach for the enhancement of speech signals corrupted by additive white noise of Gaussian statistics. The speech enhancement problem is treated as a signal estimation problem within a Bayesian framework. The conventional all-pole speech production model is assumed to govern the behaviour of the clean speech signal. The additive noise level and all-pole model gain are automatically inferred during the speech enhancement process. The strength of the Bayesian approach developed in this paper lies in its ability to perform speech enhancement without the usual requirement of estimating the level of the corrupting noise from “silence” segments of the corrupted signal. The performance of the Baycsian approach is compared to that of the Lim B Oppenheim framework, IO which it follows a similar iterative nature. A significant quality improvement is obtained over the Lim B Oppenheim framework. all-pole model are related to hyperparumerers within the Bayesian framework [l3, 15. 141. The optimisation of these hyperparameters within an iterative framework leads to an estimate of the clean speech signal and the coefficients of the all-pole, or equivalently Linear Prediction (LP), model.