AUDIO SOURCE SEPARATION INFORMED BY REDUNDANCY WITH GREEDY MULTISCALE DECOMPOSITIONS

Abstract :

This paper describes a greedy algorithm for audio source separation of repeated musical patterns. The problem is understood as retrieving from a set of mixtures the part that is redundant among them and the parts that are specific to only one mixture. The key assumption is the sparsity of all the sources in the same multiscale dictionary. Synthetic and real life examples of source separation of hand cut repeated musical patterns are exposed. Results shows that the proposed method succeeds in simultaneously providing a sparse approximant of the mixtures and a separation of the sources.

Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal-imt.archives-ouvertes.fr/hal-00735234
Contributor : Admin Télécom Paristech <>
Submitted on : Wednesday, September 26, 2012 - 3:11:35 PM
Last modification on : Thursday, October 17, 2019 - 12:37:02 PM
Long-term archiving on : Thursday, December 27, 2012 - 3:51:49 AM

File

Moussallam_Richard_Daudet_-_20...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00735234, version 1

Collections

Citation

Manuel Moussallam, Gael Richard, Laurent Daudet. AUDIO SOURCE SEPARATION INFORMED BY REDUNDANCY WITH GREEDY MULTISCALE DECOMPOSITIONS. European Signal Processing Conference, Aug 2012, Bucarest, Romania. pp.2644-2648. ⟨hal-00735234⟩

Share

Metrics

Record views

349

Files downloads

117