Bootstrapping deep music separation from primitive auditory grouping principles

TL;DR

This paper introduces a novel method for training deep music separation models using primitive auditory cues and unlabeled YouTube music, eliminating the need for ground truth isolated sources or synthetic mixtures.

Contribution

It proposes a bootstrapping approach that leverages primitive auditory cues to train deep networks for music source separation without ground truth data.

Findings

Successfully trained on unlabeled YouTube music recordings

Achieved effective separation of vocals from accompaniment

Demonstrated potential for unsupervised deep learning in audio separation

Abstract

Separating an audio scene such as a cocktail party into constituent, meaningful components is a core task in computer audition. Deep networks are the state-of-the-art approach. They are trained on synthetic mixtures of audio made from isolated sound source recordings so that ground truth for the separation is known. However, the vast majority of available audio is not isolated. The brain uses primitive cues that are independent of the characteristics of any particular sound source to perform an initial segmentation of the audio scene. We present a method for bootstrapping a deep model for music source separation without ground truth by using multiple primitive cues. We apply our method to train a network on a large set of unlabeled music recordings from YouTube to separate vocals from accompaniment without the need for ground truth isolated sources or artificial training mixtures.