Efficient Bayesian inference for harmonic models via adaptive posterior factorization

Emmanuel Vincent 1 Mark Plumbley 2
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Harmonic sinusoidal models are an essential tool for audio signal analysis. Bayesian harmonic models are particularly interesting, since they allow the joint exploitation of various priors on the model parameters. However existing inference methods often rely on specific prior distributions and remain rather slow for realistic data. In this article, we investigate a generic inference method based on approximate factorization of the joint posterior into a product of independent distributions on small subsets of parameters. We discuss the conditions under which this factorization holds true and propose two criteria to choose these subsets adaptively. We evaluate the resulting performance experimentally for the task of musical score transcription using different levels of factorization.
Liste complète des métadonnées

https://hal.inria.fr/inria-00142935
Contributor : Emmanuel Vincent <>
Submitted on : Friday, December 10, 2010 - 4:25:57 PM
Last modification on : Thursday, March 21, 2019 - 2:20:41 PM
Document(s) archivé(s) le : Friday, March 11, 2011 - 4:03:21 AM

File

techreport_warning.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00142935, version 2

Citation

Emmanuel Vincent, Mark Plumbley. Efficient Bayesian inference for harmonic models via adaptive posterior factorization. [Research Report] PI 1841, 2007, pp.20. ⟨inria-00142935v2⟩

Share

Metrics

Record views

291

Files downloads

114