Gabor frames and deep scattering networks in audio processing

Simple item page
dc.contributor.authorBammer, Roswithacs
dc.contributor.authorDörfler, Monikacs
dc.contributor.authorHarár, Pavolcs
dc.coverage.issue4cs
dc.coverage.volume8cs
dc.date.accessioned2020-08-18T10:57:48Z
dc.date.available2020-08-18T10:57:48Z
dc.date.issued2019-09-26cs
dc.description.abstractThis paper introduces Gabor scattering, a feature extractor based on Gabor frames and Mallat's scattering transform. By using a simple signal model for audio signals specific properties of Gabor scattering are studied. It is shown that for each layer, specific invariances to certain signal characteristics occur. Furthermore, deformation stability of the coefficient vector generated by the feature extractor is derived by using a decoupling technique which exploits the contractivity of general scattering networks. Deformations are introduced as changes in spectral shape and frequency modulation. The theoretical results are illustrated by numerical examples and experiments. Numerical evidence is given by evaluation on a synthetic and a "real" data set, that the invariances encoded by the Gabor scattering transform lead to higher performance in comparison with just using Gabor transform, especially when few training samples are available.en
dc.formattextcs
dc.format.extent1-25cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationAxioms. 2019, vol. 8, issue 4, p. 1-25.en
dc.identifier.doi10.3390/axioms8040106cs
dc.identifier.issn2075-1680cs
dc.identifier.other159057cs
dc.identifier.urihttp://hdl.handle.net/11012/194791
dc.language.isoencs
dc.publisherMDPIcs
dc.relation.ispartofAxiomscs
dc.relation.urihttps://www.mdpi.com/2075-1680/8/4/106cs
dc.rightsCreative Commons Attribution 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/2075-1680/cs
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/cs
dc.subjectmachine learningen
dc.subjectscattering transformen
dc.subjectGabor transformen
dc.subjectdeep learningen
dc.subjecttime-frequency analysisen
dc.subjectCNNen
dc.titleGabor frames and deep scattering networks in audio processingen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
sync.item.dbidVAV-159057en
sync.item.dbtypeVAVen
sync.item.insts2020.08.18 12:57:47en
sync.item.modts2020.08.18 12:15:28en
thesis.grantorVysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. Ústav telekomunikacícs
thesis.grantorVysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. oddělení-TKO-SIXcs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
axioms0800106v2.pdf
Size:
1.88 MB
Format:
Adobe Portable Document Format
Description:
axioms0800106v2.pdf
Download
Collections
Ústav telekomunikací
oddělení-TKO-SIX