Enregistré dans:
Détails bibliographiques
Auteurs principaux: Jeon, Chang-Bin, Wichern, Gordon, Germain, François G., Roux, Jonathan Le
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2402.18407
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929258644897792
author Jeon, Chang-Bin
Wichern, Gordon
Germain, François G.
Roux, Jonathan Le
author_facet Jeon, Chang-Bin
Wichern, Gordon
Germain, François G.
Roux, Jonathan Le
contents In music source separation, a standard training data augmentation procedure is to create new training samples by randomly combining instrument stems from different songs. These random mixes have mismatched characteristics compared to real music, e.g., the different stems do not have consistent beat or tonality, resulting in a cacophony. In this work, we investigate why random mixing is effective when training a state-of-the-art music source separation model in spite of the apparent distribution shift it creates. Additionally, we examine why performance levels off despite potentially limitless combinations, and examine the sensitivity of music source separation performance to differences in beat and tonality of the instrumental sources in a mixture.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18407
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Why does music source separation benefit from cacophony?
Jeon, Chang-Bin
Wichern, Gordon
Germain, François G.
Roux, Jonathan Le
Audio and Speech Processing
In music source separation, a standard training data augmentation procedure is to create new training samples by randomly combining instrument stems from different songs. These random mixes have mismatched characteristics compared to real music, e.g., the different stems do not have consistent beat or tonality, resulting in a cacophony. In this work, we investigate why random mixing is effective when training a state-of-the-art music source separation model in spite of the apparent distribution shift it creates. Additionally, we examine why performance levels off despite potentially limitless combinations, and examine the sensitivity of music source separation performance to differences in beat and tonality of the instrumental sources in a mixture.
title Why does music source separation benefit from cacophony?
topic Audio and Speech Processing
url https://arxiv.org/abs/2402.18407