Change search
ReferencesLink to record
Permanent link

Direct link
Adaptation of Bayesian models for single-channel source separation and its application to voice/music separation in popular songs
KTH, School of Electrical Engineering (EES), Sound and Image Processing.
2007 (English)In: IEEE Transactions on Audio, Speech, and Language Processing, ISSN 1558-7916, Vol. 15, no 5, 1564-1578 p.Article in journal (Refereed) Published
Abstract [en]

Probabilistic approaches can offer satisfactory solutions to source separation with a single channel, provided that the models of the sources match accurately the statistical properties of the mixed signals. However, it is not always possible to train such models. To overcome this problem, we propose to resort to an adaptation scheme for adjusting the source models with respect to the actual properties of the signals observed in the mix. In this paper; we introduce a general formalism for source model-adaptation which is expressed in the framework of Bayesian models. Particular cases of the proposed approach are then investigated experimentally on the problem of separating voice from music in popular songs. The obtained results show that an adaptation scheme can improve consistently and significantly the separation performance in comparison with nonadapted models.

Place, publisher, year, edition, pages
2007. Vol. 15, no 5, 1564-1578 p.
Keyword [en]
adaptive Wiener filtering, Bayesian model, expectation maximization (EM), Gaussian mixture model (GMM), maximum a posteriori (MAP), model adaptation, single-channel source separation, time-frequency masking
National Category
Engineering and Technology
URN: urn:nbn:se:kth:diva-37113DOI: 10.1109/TASL.2007.899291ISI: 000247547000007ScopusID: 2-s2.0-51449094735OAI: diva2:432144
Available from: 2011-08-01 Created: 2011-08-01 Last updated: 2011-08-01Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Ozerov, Alexey
By organisation
Sound and Image Processing
In the same journal
IEEE Transactions on Audio, Speech, and Language Processing
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 22 hits
ReferencesLink to record
Permanent link

Direct link