Conference Papers Year : 2022

Transfer Learning with Jukebox for Music Source Separation

Oliver Tautz
  • Function : Author
  • PersonId : 1406780
Helge Ritter
  • Function : Author
  • PersonId : 1008055
Andrew Melnik
  • Function : Author
  • PersonId : 1406781

Abstract

In this work, we demonstrate how a publicly available, pre-trained Jukebox model can be adapted for the problem of audio source separation from a single mixed audio channel. Our neural network architecture, which is using transfer learning, is quick to train and the results demonstrate performance comparable to other state-of-the-art approaches that require a lot more compute resources, training data, and time. We provide an open-source code implementation of our architecture ( https://github.com/wzaielamri/unmix ).
Fichier principal
Vignette du fichier
534967_1_En_35_Chapter.pdf (662.03 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04668675 , version 1 (07-08-2024)

Licence

Identifiers

Cite

Wadhah Zai El Amri, Oliver Tautz, Helge Ritter, Andrew Melnik. Transfer Learning with Jukebox for Music Source Separation. 18th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Jun 2022, Hersonissos, Greece. pp.426-433, ⟨10.1007/978-3-031-08337-2_35⟩. ⟨hal-04668675⟩
36 View
1 Download

Altmetric

Share

More