Optimum post-filter estimation for noise reduction in multichannel speech processing

Leukimmiatis, S; Maragos, P

dc.contributor.author	Leukimmiatis, S	en
dc.contributor.author	Maragos, P	en
dc.date.accessioned	2014-03-01T02:50:49Z
dc.date.available	2014-03-01T02:50:49Z
dc.date.issued	2006	en
dc.identifier.issn	22195491	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/35157
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-84862626687&partnerID=40&md5=07dd3e2c789fee4881319b49be1786e0	en
dc.subject.other	Area ratios	en
dc.subject.other	Beam formers	en
dc.subject.other	Cross-spectral densities	en
dc.subject.other	Estimation schemes	en
dc.subject.other	Input signal	en
dc.subject.other	Multi-channel	en
dc.subject.other	Multi-channel recording	en
dc.subject.other	Multichannel noise	en
dc.subject.other	Noise fields	en
dc.subject.other	Noise power spectrum	en
dc.subject.other	Post-filtering	en
dc.subject.other	Postfilters	en
dc.subject.other	Signal degradation	en
dc.subject.other	Acoustic noise measurement	en
dc.subject.other	Degradation	en
dc.subject.other	Estimation	en
dc.subject.other	Optimization	en
dc.subject.other	Signal processing	en
dc.subject.other	Speech processing	en
dc.subject.other	Signal to noise ratio	en
dc.title	Optimum post-filter estimation for noise reduction in multichannel speech processing	en
heal.type	conferenceItem	en
heal.publicationDate	2006	en
heal.abstract	This paper proposes a post-filtering estimation scheme for multichannel noise reduction. The proposed method is an extension and improvement of the existing Zelinski and McCowan post-filters which use the auto- and cross-spectral densities of the multichannel input signals to estimate the transfer function of the Wiener postfilter. A drawback in previous two post-filters is that the noise power spectrum at the beamformer's output is over-estimated and therefore the derived filters are sub-optimal in the Wiener sense. The proposed method overcomes this problem and can be used for the construction of an optimal post-filter which is also appropriate for a variety of different noise fields. In experiments with real noise multichannel recordings the proposed technique has shown to obtain a significant gain over the other studied methods in terms of signal-to-noise ratio, log area ratio distance and speech degradation measure. In particular the proposed post-filter presents a relative SNR enhancement of 17.3% and a relative decrease on signal degradation of 21.7% compared to the best of all the other studied methods.	en
heal.journalName	European Signal Processing Conference	en