Beat Tracking: Is 441 kHz Really Needed?

Simple item page
but.event.date25.04.2023cs
but.event.titleSTUDENT EEICT 2023cs
dc.contributor.authorIštvánek, Matěj
dc.contributor.authorMiklánek, Štěpán
dc.date.accessioned2023-07-17T05:57:35Z
dc.date.available2023-07-17T05:57:35Z
dc.date.issued2023cs
dc.description.abstractBeat tracking is essential in music informationretrieval, with applications ranging from music analysis and automaticplaylist generation to beat-synchronized effects. In recentyears, deep learning methods, usually inspired by well-knownarchitectures, outperformed other beat tracking algorithms. Thecurrent state-of-the-art offline beat tracking systems utilize temporalconvolutional and recurrent networks. Most systems use aninput sampling rate of 44.1 kHz. In this paper, we retrain multipleversions of state-of-the-art temporal convolutional networks withdifferent input sampling rates while keeping the time resolutionby changing the frame size parameter. Furthermore, we evaluateall models using standard metrics. As the main contribution,we show that decreasing the input audio recording samplingfrequency up to 5 kHz preserves most of the accuracy, and insome cases, even slightly outperforms the standard approach.en
dc.formattextcs
dc.format.extent227-231cs
dc.format.mimetypeapplication/pdfen
dc.identifier.citationProceedings II of the 29st Conference STUDENT EEICT 2023: Selected papers. s. 227-231. ISBN 978-80-214-6154-3cs
dc.identifier.doi10.13164/eeict.2023.227
dc.identifier.isbn978-80-214-6154-3
dc.identifier.issn2788-1334
dc.identifier.urihttp://hdl.handle.net/11012/210696
dc.language.isoencs
dc.publisherVysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologiícs
dc.relation.ispartofProceedings II of the 29st Conference STUDENT EEICT 2023: Selected papersen
dc.relation.urihttps://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_2_v2.pdfcs
dc.rights© Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologiícs
dc.rights.accessopenAccessen
dc.subjectBeat trackingen
dc.subjectmusic information retrievalen
dc.subjecttemporalconvolutional networksen
dc.subjectmachine learningen
dc.titleBeat Tracking: Is 441 kHz Really Needed?en
dc.type.driverconferenceObjecten
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.affiliatedInstitution.departmentFakulta elektrotechniky a komunikačních technologiícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
227_EEICT_selected.pdf
Size:
1.07 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Student EEICT 2023: Selected Papers