Ambiguity modelling with label distribution learning for music classification

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Buisson, Morgan
• dc.contributor.author Alonso-Jiménez, Pablo
• dc.contributor.author Bogdanov, Dmitry
• dc.date.accessioned 2022-05-30T06:32:04Z
• dc.date.available 2022-05-30T06:32:04Z
• dc.date.issued 2022
• dc.description Comunicació presentada a: 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), celebrat del 22 al 27 de maig de 2022 a Singapur.
• dc.description.abstract An important amount of work has been devoted to the task of music classification. Despite promising results achieved by convolutional neural networks, there still exists a gap left to be filled for such models to perform well in real-world applications. In this work, we address the issue of ambiguity that can arise in many classification problems. We propose a method based on adaptive label smoothing that aims at implicitly modelling perceptual vagueness among classes to improve both training and testing performances. We assess our method using two state-of-the-art CNN architectures for audio classification on a variety of music mood and genre classification tasks. We show that the proposed strategy brings consistent improvements over the traditional approach, significantly improves generalization to external audio collections and emphasizes how crucial information carried by labels can be in an ambiguous music classification context.
• dc.format.mimetype application/pdf
• dc.identifier.citation Buisson M, Alonso-Jimenez P, Bogdanov D. Ambiguity modelling with label distribution learning for music classification. In: 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP); 2022 May 22-27; Singapore. [New Jersery]: The Institute of Electrical and Electronics Engineers; 2022. p. 611-5. DOI: 10.1109/ICASSP43922.2022.9747467
• dc.identifier.doi http://doi.org/10.1109/ICASSP43922.2022.9747467
• dc.identifier.uri http://hdl.handle.net/10230/53294
• dc.language.iso eng
• dc.publisher Institute of Electrical and Electronics Engineers (IEEE)
• dc.relation.ispartof 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP); 2022 May 22-27; Singapore. [New Jersery]: The Institute of Electrical and Electronics Engineers; 2022. p. 611-5.
• dc.rights © 2022 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. http://dx.doi.org/10.1109/ICASSP43922.2022.9747467
• dc.rights.accessRights info:eu-repo/semantics/openAccess
• dc.subject.keyword Music classification
• dc.subject.keyword Deep learning
• dc.subject.keyword Label distribution learning
• dc.subject.keyword Generalization
• dc.subject.keyword Buisson, Morgan
• dc.title Ambiguity modelling with label distribution learning for music classification
• dc.type info:eu-repo/semantics/conferenceObject
• dc.type.version info:eu-repo/semantics/acceptedVersion