Improving music source separation based on deep neural networks through data augmentation and network blending
Stefan Uhlich, Marcello Porcu, Franck Giron, Michael Enenkl, Thomas Kemp, Naoya Takahashi, Yuki Mitsufuji
Abstract
This paper deals with the separation of music into individual instrument tracks which is known to be a challenging problem. We describe two different deep neural network architectures for this task, a feed-forward and a recurrent one, and show that each of them yields themselves state-of-the art results on the SiSEC DSD100 dataset. For the recurrent network, we use data augmentation during training and show that even simple separation networks are prone to overfitting if no data augmentation is used. Furthermore, we propose a blending of both neural network systems where we linearly combine their raw outputs and then perform a multi-channel Wiener filter post-processing. This blending scheme yields the best results that have been reported to-date on the SiSEC DSD100 dataset.
BibTeX
@inproceedings{icassp2017_improvingmusicso,
title = {Improving music source separation based on deep neural networks through data augmentation and network blending},
author = {Stefan Uhlich and Marcello Porcu and Franck Giron and Michael Enenkl and Thomas Kemp and Naoya Takahashi and Yuki Mitsufuji},
booktitle = {ICASSP 2017},
year = {2017}
}