Improved singing voice separation with chromagram-based pitch-aware remixing
This work addresses data scarcity in music separation for audio processing applications, but it is incremental as it builds on existing data augmentation methods.
The paper tackled the problem of limited training data for singing voice separation by proposing a chromagram-based pitch-aware remixing data augmentation technique, which improved the test signal-to-distortion ratio (SDR) in supervised and semi-supervised settings.
Singing voice separation aims to separate music into vocals and accompaniment components. One of the major constraints for the task is the limited amount of training data with separated vocals. Data augmentation techniques such as random source mixing have been shown to make better use of existing data and mildly improve model performance. We propose a novel data augmentation technique, chromagram-based pitch-aware remixing, where music segments with high pitch alignment are mixed. By performing controlled experiments in both supervised and semi-supervised settings, we demonstrate that training models with pitch-aware remixing significantly improves the test signal-to-distortion ratio (SDR)