Show simple item record

dc.contributor.authorVianna Lordelo, C
dc.contributor.authorBenetos, E
dc.contributor.authorDixon, S
dc.contributor.authorAhlbäck, S
dc.contributor.authorOhlsson, P
dc.date.accessioned2021-01-06T14:24:18Z
dc.date.available2020-12-11
dc.date.available2021-01-06T14:24:18Z
dc.date.issued2020
dc.identifier.issn1070-9908
dc.identifier.urihttps://qmro.qmul.ac.uk/xmlui/handle/123456789/69527
dc.description.abstractThis paper addresses the problem of domain adaptation for the task of music source separation. Using datasets from two different domains, we compare the performance of a deep learning-based harmonic-percussive source separation model under different training scenarios, including supervised joint training using data from both domains and pre-training in one domain with fine-tuning in another. We propose an adversarial unsupervised domain adaptation approach suitable for the case where no labelled data (ground-truth source signals) from a target domain is available. By leveraging unlabelled data (only mixtures) from this domain, experiments show that our framework can improve separation performance on the new domain without losing any considerable performance on the original domain. The paper also introduces the Tap & Fiddle dataset, a dataset containing recordings of Scandinavian fiddle tunes along with isolated tracks for "foot-tapping" and "violin".en_US
dc.format.extent? - ? (5)
dc.publisherInstitute of Electrical and Electronics Engineersen_US
dc.relation.ispartofIEEE Signal Processing Letters
dc.titleAdversarial Unsupervised Domain Adaptation for Harmonic-Percussive Source Separationen_US
dc.typeArticleen_US
dc.rights.holder© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
pubs.notesNot knownen_US
pubs.publication-statusAccepteden_US
dcterms.dateAccepted2020-12-11
rioxxterms.funderDefault funderen_US
rioxxterms.identifier.projectDefault projecten_US
qmul.funderNew Frontiers in Music Information Processing (MIP-Frontiers)::European Commissionen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record