Real-time detection of overlapping sound events with non-negative matrix factorization

Arnaud Dessein 1, 2 Arshia Cont 1, 3 Guillaume Lemaitre 2
1 MuTant - Synchronous Realtime Processing and Programming of Music Signals
Inria Paris-Rocquencourt, UPMC - Université Pierre et Marie Curie - Paris 6, IRCAM, CNRS - Centre National de la Recherche Scientifique
3 Musical Representations
STMS - Sciences et Technologies de la Musique et du Son
Abstract : In this paper, we investigate the problem of real-time detection of overlapping sound events by employing non-negative matrix factorization techniques. We consider a setup where audio streams arrive in real-time to the system and are decomposed onto a dictionary of event templates learned off-line prior to the decomposition. An important drawback of existing approaches in this context is the lack of controls on the decomposition. We propose and compare two provably convergent algorithms that address this issue, by controlling respectively the sparsity of the decomposition and the trade-off of the decomposition between the different frequency components. Sparsity regularization is considered in the framework of convex quadratic programming, while frequency compromise is introduced by employing the beta-divergence as a cost function. The two algorithms are evaluated on the multi-source detection tasks of polyphonic music transcription, drum transcription and environmental sound recognition. The obtained results show how the proposed approaches can improve detection in such applications, while maintaining low computational costs that are suitable for real-time.
Type de document :
Chapitre d'ouvrage
Nielsen, Frank and Bhatia, Rajendra. Matrix Information Geometry, Springer, pp.341-371, 2013, 978-3-642-30232-9. 〈10.1007/978-3-642-30232-9_14〉
Liste complète des métadonnées

Littérature citée [59 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00708805
Contributeur : Arnaud Dessein <>
Soumis le : vendredi 15 juin 2012 - 17:41:18
Dernière modification le : vendredi 31 août 2018 - 09:14:29
Document(s) archivé(s) le : dimanche 16 septembre 2012 - 03:01:15

Fichier

Dessein2012MIG.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Arnaud Dessein, Arshia Cont, Guillaume Lemaitre. Real-time detection of overlapping sound events with non-negative matrix factorization. Nielsen, Frank and Bhatia, Rajendra. Matrix Information Geometry, Springer, pp.341-371, 2013, 978-3-642-30232-9. 〈10.1007/978-3-642-30232-9_14〉. 〈hal-00708805〉

Partager

Métriques

Consultations de la notice

516

Téléchargements de fichiers

439