Perceptually controlled doping for audio source separation

Abstract : The separation of an underdetermined audio mixture can be performed through sparse component analysis (SCA) that relies however on the strong hypothesis that source signals are sparse in some domain. To overcome this difficulty in the case where the original sources are available before the mixing process, the informed source separation (ISS) embeds in the mixture a watermark, which information can help a further separation. Though powerful, this technique is generally specific to a particular mixing setup and may be compromised by an additional bitrate compression stage. Thus, instead of watermarking, we propose a 'doping' method that makes the time-frequency representation of each source more sparse, while preserving its audio quality. This method is based on an iterative decrease of the distance between the distribution of the signal and a target sparse distribution, under a perceptual constraint. We aim to show that the proposed approach is robust to audio coding and that the use of the sparsified signals improves the source separation, in comparison with the original sources. In this work, the analysis is made only in instantaneous mixtures and focused on voice sources.
Type de document :
Article dans une revue
Liste complète des métadonnées

Littérature citée [43 références]  Voir  Masquer  Télécharger

https://hal-descartes.archives-ouvertes.fr/hal-01839081
Contributeur : Gaël Mahé <>
Soumis le : vendredi 13 juillet 2018 - 17:33:44
Dernière modification le : vendredi 12 avril 2019 - 11:14:05
Archivage à long terme le : lundi 15 octobre 2018 - 13:30:34

Fichier

10.1186%2F1687-6180-2014-27.pd...
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

  • HAL Id : hal-01839081, version 1

Collections

Citation

Gaël Mahé, Everton Nadalin, Ricardo Suyama, Joao Romano. Perceptually controlled doping for audio source separation. EURASIP Journal on Advances in Signal Processing, SpringerOpen, 2014, 2014, pp.27 - 27. ⟨hal-01839081⟩

Partager

Métriques

Consultations de la notice

65

Téléchargements de fichiers

133