A General Variational Bayesian Framework for Robust Feature Extraction in Multisource Recordings

Kamil Adiloglu 1 Emmanuel Vincent 1
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : We consider the problem of extracting features from individual sources in a multisource audio recording using a general source separation algorithm. The main issue is to estimate and propagate the uncertainty over the separated source signals, so as to robustly estimate the features despite source separation errors. While state-of-the-art techniques estimate the uncertainty in a heuristic manner, we propose to integrate over the parameter space of the source separation algorithm. We apply variational Bayes to estimate the posterior probability of the sources and subsequently derive the expectation of the features by moment matching. Experiments over stereo mixtures of three or four sources show that the proposed method provides the best results in terms of the root mean square (RMS) error on the estimated features.
Liste complète des métadonnées

https://hal.inria.fr/hal-00656613
Contributor : Kamil Adiloglu <>
Submitted on : Wednesday, January 4, 2012 - 3:51:28 PM
Last modification on : Thursday, March 21, 2019 - 2:20:42 PM
Document(s) archivé(s) le : Monday, November 19, 2012 - 12:20:24 PM

File

VARNMF.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00656613, version 1

Citation

Kamil Adiloglu, Emmanuel Vincent. A General Variational Bayesian Framework for Robust Feature Extraction in Multisource Recordings. IEEE International Conference on Acoustics, Speech and Signal Processing, Mar 2012, Kyoto, Japan. 2012. 〈hal-00656613〉

Share

Metrics

Record views

464

Files downloads

319