Technical Program

Paper Detail

Paper:	ITT-L1.3
Session:	Speech Processing Applications
Time:	Thursday, May 18, 17:10 - 17:30
Presentation:	Lecture
Topic:	Industry Technology Track: Speech Enhancement
Title:	NOISE POWER SPECTRUM ESTIMATION FOR SPEECH ENHANCEMENT USING AN AUTOREGRESSIVE MODEL FOR SPEECH POWER SPECTRUM DYNAMICS
Authors:	Ivo Batina, Jesper Jensen, Richard Heusdens, Technical University of Delft, Netherlands
Abstract:	In this paper we propose a method for estimating the non-stationary noise power spectral density (PSD) given a noisy speech signal. The method is based on an autoregressive (AR) model of the speech PSD dynamics combined with a Kalman filtering based noise PSD estimation technique. Objective and subjective performance evaluations show that the speech enhancement scheme utilizing the proposed noise PSD estimation technique achieves significant improvements over a system using a stationary noise estimate as well as compared to a system that uses a noise tracker developed in our previous work.