Tampere University of Technology

TUTCRIS Research Portal

Towards Complex Nonnegative Matrix Factorization with the Beta-Divergence

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Details

Original languageEnglish
Title of host publication2018 16th International Workshop on Acoustic Signal Enhancement (IWAENC)
PublisherIEEE
Pages156-160
ISBN (Electronic)978-1-5386-8151-0
ISBN (Print)978-1-5386-8152-7
DOIs
Publication statusPublished - Sep 2018
Publication typeA4 Article in a conference publication
EventInternational Workshop on Acoustic Signal Enhancement -
Duration: 1 Jan 1900 → …

Conference

ConferenceInternational Workshop on Acoustic Signal Enhancement
Period1/01/00 → …

Abstract

Complex nonnegative matrix factorization (NMF) is a powerful tool for decomposing audio spectrograms while accounting for some phase information in the time-frequency domain. While its estimation was originally based on the Euclidean distance, in this paper we propose to extend it to any beta-divergence, a family of functions widely used in audio to estimate NMF. To this end, we introduce the beta-divergence in a heuristic fashion within a phase-aware probabilistic model. Estimating this model results in performing an NMF with Itakura-Saito (IS) divergence on a quantity called the phase-corrected posterior power of the sources, which is both phase-dependent and nonnegative-valued. Therefore, we replace IS with the beta-divergence, so that the factorization uses an optimal distortion metric and remains phase-aware. Even though by doing so we loose theoretical convergence guarantees, the resulting algorithm demonstrates its potential for an audio source separation task, where it outperforms previous complex NMFs approaches.

Publication forum classification

Field of science, Statistics Finland