Less is more: faster and better music version identification with embedding distillation

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Serrà Julià, Joan
• dc.contributor.author Yesiler, Furkan
• dc.contributor.author Gómez Gutiérrez, Emilia, 1975-
• dc.date.accessioned 2020-11-11T08:43:28Z
• dc.date.available 2020-11-11T08:43:28Z
• dc.date.issued 2020
• dc.description Comunicació presentada a: International Society for Music Information Retrieval Conference celebrat de l'11 al 16 d'octubre de 2020 de manera virtual.
• dc.description.abstract Version identification systems aim to detect different renditions of the same underlying musical composition (loosely called cover songs). By learning to encode entire recordings into plain vector embeddings, recent systems have made significant progress in bridging the gap between accuracy and scalability, which has been a key challenge for nearly two decades. In this work, we propose to further narrow this gap by employing a set of data distillation techniques that reduce the embedding dimensionality of a pre-trained state-of-the-art model. We compare a wide range of techniques and propose new ones, from classical dimensionality reduction to more sophisticated distillation schemes. With those, we obtain 99% smaller embeddings that, moreover, yield up to a 3% accuracy increase. Such small embeddings can have an important impact in retrieval time, up to the point of making a real-world system practical on a standalone laptop.en
• dc.description.sponsorship This work is supported by the MIP-Frontiers project, the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 765068, and by TROMPA, the Horizon 2020 project 770376-2.
• dc.format.mimetype application/pdf
• dc.identifier.citation Yesiler F, Serrà J, Gómez E. Less is more: faster and better music version identification with embedding distillation. In: Cumming J, Ha Lee J, McFee B, Schedl M, Devaney J, McKay C, Zagerle E, de Reuse T, editors. Proceedings of the 21st International Society for Music Information Retrieval Conference; 2020 Oct 11-16; Montréal, Canada. [Canada]: ISMIR; 2020. p. 884-92.
• dc.identifier.uri http://hdl.handle.net/10230/45718
• dc.language.iso eng
• dc.publisher International Society for Music Information Retrieval (ISMIR)
• dc.relation.ispartof Cumming J, Ha Lee J, McFee B, Schedl M, Devaney J, McKay C, Zagerle E, de Reuse T, editors. Proceedings of the 21st International Society for Music Information Retrieval Conference; 2020 Oct 11-16; Montréal, Canada. [Canada]: ISMIR; 2020. p. 884-92
• dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/765068
• dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/770376-2
• dc.rights © F. Yesiler, J. Serrà and E. Gómez. Licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0). Attribution: F. Yesiler, J. Serrà and E. Gómez, “Less is more: Faster and better music version identification with embedding distillation”, in Proc. of the 21st Int. Society for Music Information Retrieval Conf., Montréal, Canada, 2020.
• dc.rights.accessRights info:eu-repo/semantics/openAccess
• dc.rights.uri https://creativecommons.org/licenses/by/4.0/
• dc.title Less is more: faster and better music version identification with embedding distillationen
• dc.type info:eu-repo/semantics/conferenceObject
• dc.type.version info:eu-repo/semantics/publishedVersion