Towards complex matrix decomposition of spectrograms based on the relative phase offsets of harmonic sounds

Holger Kirchhoff 1 Roland Badeau 2 Simon Dixon 1
1 QMUL - Queen Mary University of London
2 LTCI - Laboratoire Traitement et Communication de l'Information
Abstract : In this paper we study the relative phase offsets between partials in the sustained part of harmonic sounds and investigate their suitability for complex matrix decomposition of spectrograms. We formally introduce this property in a sinusoidal model and visualise the phase relations of a musical instrument. A model of complex matrix decomposition in the time-frequency domain is derived and equations for the estimation of the model parameters are provided in the monophonic case. We illustrate the model with the analysis of a monophonic saxophone signal. The results suggest that the phase offset is able to capture inherent time-invariant phase properties of harmonic sounds and outline its potential use for complex matrix decomposition.
Keywords : nonnegative matrix factorisation Harmonic signals relative phase offsets of partials complex matrix decomposition nonnegative matrix factorisation.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [15 references]  Display  Hide  Download
https://hal.inria.fr/hal-00945295
Contributor : Roland Badeau <>
Submitted on : Thursday, May 15, 2014 - 6:23:18 PM
Last modification on : Wednesday, February 20, 2019 - 2:40:26 PM
Document(s) archivé(s) le : Friday, August 15, 2014 - 10:37:38 AM

File

Kirchhoff-ICASSP2014.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00945295, version 1

Collections

INSTITUT-TELECOM | PARISTECH | LTCI

Citation

Holger Kirchhoff, Roland Badeau, Simon Dixon. Towards complex matrix decomposition of spectrograms based on the relative phase offsets of harmonic sounds. Proc. of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 2014, Florence, Italy. pp.1591-1595. ⟨hal-00945295⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

256

Files downloads

433

Contact


archive-ouverte@inria.fr