Bespoke Neural Networks for Score-Informed Source Separation
Abstract
In this paper, we introduce a simple method that can separate arbitrary musical instruments from an audio mixture. Given an unaligned MIDI transcription for a target instrument from an input mixture, we synthesize new mixtures from the midi transcription that sound similar to the mixture to be separated. This lets us create a labeled training set to train a network on the specific bespoke task. When this model applied to the original mixture, we demonstrate that this method can: 1) successfully separate out the desired instrument with access to only unaligned MIDI, 2) separate arbitrary instruments, and 3) get results in a fraction of the time of existing methods. We encourage readers to listen to the demos posted here: https://git.io/JUu5q.
Cite
@article{arxiv.2009.13729,
title = {Bespoke Neural Networks for Score-Informed Source Separation},
author = {Ethan Manilow and Bryan Pardo},
journal= {arXiv preprint arXiv:2009.13729},
year = {2020}
}
Comments
ISMIR 2020 - Late Breaking Demo