GMM based Bayesian approach to speech enhancement in signal/transform domain
2008 (English)In: 2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, New York: IEEE , 2008, 4893-4896 p.Conference paper (Refereed)
Considering a general linear model of signal degradation, by modeling the probability density function (PDF) of the clean signal using a Gaussian mixture model (GMM) and additive noise by a Gaussian PDF, we derive the minimum mean square error (MMSE) estimator. The derived MMSE estimator is non-linear and the linear MMSE estimator is shown to be a special case. For speech signal corrupted by independent additive noise, by modeling the joint PDF of time-domain speech samples of a speech frame using a GMM, we propose a speech enhancement method based on the derived MMSE estimator. We also show that the same estimator can be used for transform-domain speech enhancement.
Place, publisher, year, edition, pages
New York: IEEE , 2008. 4893-4896 p.
, International Conference on Acoustics Speech and Signal Processing ICASSP, ISSN 1520-6149
Speech enhancement, estimation
Research subject SRA - ICT
IdentifiersURN: urn:nbn:se:kth:diva-46574DOI: 10.1109/ICASSP.2008.4518754ISI: 000257456703207ISBN: 978-1-4244-1483-3OAI: oai:DiVA.org:kth-46574DiVA: diva2:453899
33rd IEEE International Conference on Acoustics, Speech and Signal Processing. Las Vegas, NV. MAR 30-APR 04, 2008
QC 201111072011-11-032011-11-032012-01-20Bibliographically approved