Frequency domain multi-channel expectation maximization algorithm for audio background noise reduction

Author

Jichi Deng ; Godsill, Simon

Author_Institution

Dept. of Eng., Univ. of Cambridge, Cambridge, UK

fYear

2013

fDate

20-23 Oct. 2013

Firstpage

1

Lastpage

4

Abstract

In this paper we implement expectation maximization (EM) based methods in the short time Fourier transform (STFT) domain for background noise reduction in multi-channel systems. The models introduce a Wishart prior for the unknown signal covariance matrix. An EM algorithm is used to maximise the posterior probability for the clean signal, approaching a stationary point of the distribution with increasing iterations. The background noise is modelled as white and stationary in this initial work. The proposed methods are found to outperform a multi-channel Wiener filter in terms of residual noise artefacts and MSE for a small initial trial.

Keywords

Fourier transforms; audio signal processing; covariance matrices; expectation-maximisation algorithm; iterative methods; probability; EM algorithm; MSE; STFT domain; audio background noise reduction; background noise; expectation maximization based methods; frequency domain multichannel expectation maximization algorithm; iterations; multichannel Wiener filter; multichannel systems; posterior probability; residual noise artefacts; short time Fourier transform domain; signal covariance matrix; Noise measurement; Noise reduction; Signal processing algorithms; Time-frequency analysis; White noise; Expectation maximization; Wiener filter; maximum aposteriori; noise reduction; short time Fourier transform;

fLanguage

English

Publisher

ieee

Conference_Titel

Applications of Signal Processing to Audio and Acoustics (WASPAA), 2013 IEEE Workshop on

Conference_Location

New Paltz, NY

ISSN

1931-1168

Type

conf

DOI

10.1109/WASPAA.2013.6701859

Filename

6701859