Common Fate Model for Unison source Separation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Common Fate Model for Unison source Separation

Résumé

In this paper we present a novel source separation method aiming to overcome the difficulty of modelling non-stationary signals. The method can be applied to mixtures of musical instruments with frequency and/or amplitude modulation, e.g. typically caused by vi-brato. It is based on a signal representation that divides the complex spectrogram into a grid of patches of arbitrary size. These complex patches are then processed by a two-dimensional discrete Fourier transform, forming a tensor representation which reveals spectral and temporal modulation textures. Our representation can be seen as an alternative to modulation transforms computed on magnitude spectrograms. An adapted factorization model allows to decompose different time-varying harmonic sources based on their particular common modulation profile: hence the name Common Fate Model. The method is evaluated on musical instrument mixtures playing the same fundamental frequency (unison), showing improvement over other state-of-the-art methods.
Fichier principal
Vignette du fichier
common_fate_icassp2016.pdf (1.05 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01248012 , version 1 (27-01-2016)

Identifiants

  • HAL Id : hal-01248012 , version 1

Citer

Fabian-Robert Stöter, Antoine Liutkus, Roland Badeau, Bernd Edler, Paul Magron. Common Fate Model for Unison source Separation. 41st International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2016, Shanghai, China. ⟨hal-01248012⟩
591 Consultations
988 Téléchargements

Partager

Gmail Facebook X LinkedIn More