openBliSSART is a C++ framework and toolbox that provides "Blind Source
Separation for Audio Recognition Tasks". Its areas of application include
instrument separation (e.g. extraction of drum tracks from popular music),
speech enhancement, and feature extraction. It features various source
separation algorithms, with a strong focus on variants of Non-Negative Matrix
Factorization (NMF).

Besides basic unsupervised source separation, it provides support for component
classification by Support Vector Machines (SVM) using common acoustic features
from speech and music processing. For data set creation a Qt-based GUI is
available. Furthermore, supervised NMF can be performed and used for audio
feature extraction.
 
openBliSSART is fast: typical real-time factors are in the order of 0.1
(Euclidean NMF) on a state-of-the-art desktop PC.

openBliSSART is free software and licensed under the GNU General Public License
(see the COPYING file).

Detailed installation instructions can be found in the INSTALL file. Make sure
to read it first, as it contains valuable hints for easy installation on many
system configurations.

For a first impression of openBliSSART, you may want to try the drum beat
separation demonstrator which is included in the "demo" directory, along with
installation and usage information.

There is a variety of documentation available in the "doc" directory, including
a tutorial, reference manual, and API documentation. Please consult the file
"doc/README" for details.

If you want to use openBliSSART for your research, please cite the following
paper:

Felix Weninger, Alexander Lehmann, Bjoern Schuller: 
"openBliSSART: Design and Evaluation of a Research Toolkit for 
Blind Source Separation in Audio Recognition Tasks", to appear in
Proc. International Conference on Acoustics, Speech and Signal Processing 
(ICASSP) 2011, IEEE, Prague, Czech Republic, 22.-27.05.2011.