Matlab code of the algorithm described in the paper: R. Hennequin, B. David, R. Badeau. "Score informed audio source separation using a parametric model of non-negative spectrogram", IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), May 2011, Prague, Czech Republic
The paper is available on this page: http://www.romain-hennequin.fr/En/publications/publications.html
Follow the instruction below :
- clone the git repository on your hard drive
- open matlab and cd to the directory where you cloned the repository
- run setup.m (if asked, click on "add to path")
You should then be able to use any functions or scripts in this directory. You can test the code using the script script_SISS.m (might take a little while)
This code is released under the GPL license. It contains code by Ken Schutte (MIDI file tools). See http://www.kenschutte.com/midi for more info about code by Ken Schutte. It contains code by Dan Ellis (stft/istft). See http://www.ee.columbia.edu/~dpwe/ for more info about code Dan Ellis.