Skip to Main content Skip to Navigation
Conference papers

Factorial scaled hidden Markov model for polyphonic audio representation and source separation

Abstract : We present a new probabilistic model for polyphonic audio termed Factorial Scaled Hidden Markov Model (FS-HMM), which generalizes several existing models, notably the Gaussian scaled mixture model and the Itakura-Saito Nonnegative Matrix Factorization (NMF) model. We describe two expectation-maximization (EM) algorithms for maximum likelihood estimation, which differ by the choice of complete data set. The second EM algorithm, based on a reduced complete data set and multiplicative updates inspired from NMF methodology, exhibits much faster convergence. We consider the FS-HMM in different configurations for the difficult problem of speech / music separation from a single channel and report satisfying results.
Complete list of metadata

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/inria-00553336
Contributor : Alexey Ozerov <>
Submitted on : Friday, January 7, 2011 - 11:46:07 AM
Last modification on : Thursday, January 7, 2021 - 8:18:14 PM
Long-term archiving on: : Friday, April 8, 2011 - 2:57:34 AM

File

OzerovFevotteCharbit_WASPAA09....
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00553336, version 1

Collections

Citation

Alexey Ozerov, Cédric Févotte, Maurice Charbit. Factorial scaled hidden Markov model for polyphonic audio representation and source separation. IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA'09), Oct 2009, Mohonk, NY, United States. ⟨inria-00553336⟩

Share

Metrics

Record views

268

Files downloads

714