Nonnegative tensor factorization with frequency modulation cues for blind audio source separation

Elliot Creager, Noah D. Stein, Roland Badeau, Philippe Depalle

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

Abstract

We present Vibrato Nonnegative Tensor Factorization, an algorithm for single-channel unsupervised audio source separation with an application to separating instrumental or vocal sources with nonstationary pitch from music recordings. Our approach extends Nonnegative Matrix Factorization for audio modeling by including local estimates of frequency modulation as cues in the separation. This permits the modeling and unsupervised separation of vibrato or glissando musical sources, which is not possible with the basic matrix factorization formulation. The algorithm factorizes a sparse nonnegative tensor comprising the audio spectrogram and local frequency-slope-to-frequency ratios, which are estimated at each time-frequency bin using the Distributed Derivative Method. The use of local frequency modulations as separation cues is motivated by the principle of common fate partial grouping from Auditory Scene Analysis, which hypothesizes that each latent source in a mixture is characterized perceptually by coherent frequency and amplitude modulations shared by its component partials. We derive multiplicative factor updates by Minorization-Maximization, which guarantees convergence to a local optimum by iteration. We then compare our method to the baseline on two separation tasks: one considers synthetic vibrato notes, while the other considers vibrato string instrument recordings.

Original languageEnglish
Title of host publicationProceedings of the 17th International Society for Music Information Retrieval Conference, ISMIR 2016
EditorsMichael I. Mandel, Johanna Devaney, Douglas Turnbull, George Tzanetakis
PublisherInternational Society for Music Information Retrieval
Pages211-217
Number of pages7
ISBN (Electronic)9780692755068
Publication statusPublished - 1 Jan 2016
Externally publishedYes
Event17th International Society for Music Information Retrieval Conference, ISMIR 2016 - New York, United States
Duration: 7 Aug 201611 Aug 2016

Publication series

NameProceedings of the 17th International Society for Music Information Retrieval Conference, ISMIR 2016

Conference

Conference17th International Society for Music Information Retrieval Conference, ISMIR 2016
Country/TerritoryUnited States
CityNew York
Period7/08/1611/08/16

Fingerprint

Dive into the research topics of 'Nonnegative tensor factorization with frequency modulation cues for blind audio source separation'. Together they form a unique fingerprint.

Cite this