Change search
ReferencesLink to record
Permanent link

Direct link
A Bayesian hierarchical mixture of experts approach to estimate speech quality
KTH, School of Electrical Engineering (EES), Sound and Image Processing.
Show others and affiliations
2010 (English)In: 2010 2nd International Workshop on Quality of Multimedia Experience, 2010, Vol. QoMEX 2010 - Proceedings, 200-205 p.Conference paper (Refereed)
Abstract [en]

This paper demonstrates the potential of theoretically motivated learning methods in solving the problem of non-intrusive quality estimation for which the state-of-the-art is represented by ITU-T P.563 standard. To construct our estimator, we adopt the speech features from P.563, while we use a different mapping of features to form quality estimates. In contrast to P.563 which assumes distortion-classes to divide the feature space, our approach divides the feature space based on a clustering which is learned from the data using Bayesian inference. Despite using weaker modeling assumptions, we are still able to achieve comparable accuracy on predicting mean-opinion-scores with P.563. Our work suggests Bayesian model-evidence as an alternative metric to correlation-coefficient for determining the necessary number of experts for modeling the data.

Place, publisher, year, edition, pages
2010. Vol. QoMEX 2010 - Proceedings, 200-205 p.
Keyword [en]
Bayesian, Mixture of experts, Non-intrusive, Objective quality assessment, Output-based, P.563, Single-ended, Speech, Variational, Inference engines, Mixtures, Bayesian networks
National Category
Computer and Information Science
URN: urn:nbn:se:kth:diva-36317DOI: 10.1109/QOMEX.2010.5516203ScopusID: 2-s2.0-77955764588ISBN: 9781424469604OAI: diva2:430670
2010 2nd International Workshop on Quality of Multimedia Experience, QoMEX 2010; Trondheim
QC 20110712Available from: 2011-07-12 Created: 2011-07-11 Last updated: 2011-07-12Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Petkov, PetkoKleijn, W. Bastiaan
By organisation
Sound and Image Processing
Computer and Information Science

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: 24 hits
ReferencesLink to record
Permanent link

Direct link