An investigation of discrete-state discriminant approaches to single-sensor source separation

Valentin Emiya 1 Emmanuel Vincent 1 Rémi Gribonval 1
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : This paper investigated a new scheme for single-sensor audio source separation. This framework is introduced comparatively to the existing Gaussian mixture model generative approach and is focusing on the mixture states rather than on the source states, resulting in a discrete, joint state discriminant approach. The study establishes the theoretical performance bounds of the proposed scheme and an actual source separation system is designed. The performance is computed on a set of musical recordings and a discussion is proposed, including the question of the source correlation and the possible drawbacks of the method.
Type de document :
Communication dans un congrès
Proc. IEEE Work. Appli. Sig. Proces. Audio and Acous. (WASPAA), Oct 2009, New Paltz, NY, United States. pp.97-100, 2009, 〈10.1109/ASPAA.2009.5346515〉
Liste complète des métadonnées

Littérature citée [10 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00452636
Contributeur : Valentin Emiya <>
Soumis le : mardi 2 février 2010 - 17:20:19
Dernière modification le : mercredi 16 mai 2018 - 11:23:03
Document(s) archivé(s) le : vendredi 18 juin 2010 - 18:28:34

Fichier

2009WaspaaEmiya.pdf
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

Citation

Valentin Emiya, Emmanuel Vincent, Rémi Gribonval. An investigation of discrete-state discriminant approaches to single-sensor source separation. Proc. IEEE Work. Appli. Sig. Proces. Audio and Acous. (WASPAA), Oct 2009, New Paltz, NY, United States. pp.97-100, 2009, 〈10.1109/ASPAA.2009.5346515〉. 〈inria-00452636〉

Partager

Métriques

Consultations de la notice

639

Téléchargements de fichiers

168