Monaural score-informed source separation for classical music using convolutional neural networks

TitleMonaural score-informed source separation for classical music using convolutional neural networks
Publication TypeConference Paper
Year of Publication2017
Conference Name18th International Society for Music Information Retrieval Conference
AuthorsMiron, M., Janer J., & Gómez E.
Conference Start Date24/10/2017
Conference LocationSuzhou, China
Keywordsclassical music, deep learning, source separation
preprint/postprint documenthttps://www.researchgate.net/publication/318637038_Monaural_score-informed_source_separation_for_classical_music_using_convolutional_neural_networks
Additional material: 

This paper follows the research reproducibility principles. 

The code is available through a github repository.

We test our method with a well known classical music dataset, Bach10, which can be found online

The training data is generated from the audio samples in the RWC instrument samples dataset with the code in the github repository.

The separated tracks, the CNN trained model and the .mat files corresponding to the results in terms of SDR,SIR,SAR can be found at the zenodo repository.

intranet