Si vous êtes un personnel Inria et que vous souhaitez participer aux tests de notre future plateforme Gitlab basée sur la version ultimate avec gitlab LFS activé merci de contacter Didier Chassignol.

This repository contains versions 2.1 and 3.0 of BSS Eval previously available at http://bass-db.gforge.inria.fr/bss_eval/.

# Purpose

BSS Eval is a Matlab toolbox to **measure the performance** of (blind) source separation algorithms within an evaluation framework where the original source signals are available as ground truth [1,3. The measures are based on the decomposition of each estimated source signal into a number of contributions corresponding to the **target source**, **interference** from unwanted sources, and **artifacts** such as "musical noise". They are **valid for any type of data** (audio, biomedical, etc), **any mixture** (instantaneous, convolutive, etc) and **any algorithm** (beamforming, ICA, time-frequency masking, etc).

For audio data, the resulting energy ratio criteria correlate with subjective ratings to a certain extent only. For improved correlation with subjective ratings, try the [PEASS](https://gitlab.inria.fr/bass-db/peass) toolkit.

# Versions

[Version 3.0](v3.0)(August 22, 2011)

This version provides three complementary functions forming the core of the annual [Signal Separation Evaluation Campaign](http://sisec.wiki.irisa.fr/)[3]:

-[bss_eval_sources.m](v3.0/bss_eval_sources.m) for the evaluation of estimated single-channel source signals [1]

-[bss_eval_images.m](v3.0/bss_eval_images.m) for the evaluation of estimated multichannel spatial source images [3,4]

-[bss_eval_mix.m](v3.0/bss_eval_mix.m) for the evaluation of estimated mixing filters [3,5]

[Version 2.1](v2.1)(July 21, 2008)

This version was developed with the support of [GDR ISIS](http://gdr-isis.fr). It provides detailed evaluation capabilities (distinction between additive noise and other interfering sources, time-varying performance metrics) [1,2] but is practically restricted to single-channel source signals within instantaneous mixtures. For convolutive mixtures, the use of [version 3.0](v3.0) is recommended due to a greater computational efficiency enabling longer decomposition filters hence better correlation with subjective ratings.

# Usage

For examples of use of Version 3.0, see [example_inst.m](http://sisec2008.wiki.irisa.fr/example_inst0a77.m?fileId=13) and [example_conv.m](http://sisec2008.wiki.irisa.fr/example_conv69a4.m?fileId=14) on the [Signal Separation Evaluation Campaign](http://sisec.wiki.irisa.fr/) website.

For usage instructions of Version 2.1, see the [user guide](v2.1/user_guide.pdf).

# References

[1] E. Vincent, R. Gribonval, and C. Févotte, ["Performance measurement in blind audio source separation"](https://hal.inria.fr/inria-00544230/document), *IEEE Transactions on Audio, Speech, and Language Processing*, 14(4), pp 1462-1469, 2006.

[2] C. Févotte, R. Gribonval, and E. Vincent, [BSS_EVAL toolbox user guide - Revision 2.0](https://hal.inria.fr/inria-00564760/document), Technical Report 1706, IRISA, April 2005.

[3] E. Vincent, S. Araki, F.J. Theis, G. Nolte, P. Bofill, H. Sawada, A. Ozerov, B.V. Gowreesunker, D. Lutter, and N.Q.K. Duong, ["The Signal Separation Evaluation Campaign (2007-2010): Achievements and remaining challenges"](https://hal.inria.fr/inria-00630985/document), *Signal Processing*, 92, pp. 1928-1936, 2012.

[4] E. Vincent, H. Sawada, P. Bofill, S. Makino, and J.P. Rosca, ["First stereo audio source separation evaluation campaign: Data, algorithms and results"](https://hal.inria.fr/inria-00544199/document), in *Proc. 7th Int. Conf. on Independent Component Analysis and Signal Separation (ICA)*, pp 552-559, 2007.

[5] E. Vincent, S. Araki, and P. Bofill, ["The 2008 Signal Separation Evaluation Campaign: A community-based approach to large-scale evaluation"](https://hal.inria.fr/inria-00544168/document), in *Proc. 8th Int. Conf. on Independent Component Analysis and Signal Separation (ICA)*, pp 734-741, 2009.

# License

Version 3.0 was authored by Emmanuel Vincent and is released under the terms of the [GNU General Public License (GPL) version 3](https://www.gnu.org/licenses/gpl-3.0.en.html).

Version 2.1 was authored by Cédric Févotte, Rémi Gribonval and Emmanuel Vincent and is released under the terms of the [GNU General Public License (GPL) version 2](https://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html).

% bss_crit - compute evaluation criteria given a decomposition of an estimated source into target/interference/noise/artifacts

% bss_decomp_filt - decompose an estimated source into target/interference/noise/artefacts components, assuming the admissible distortion is a pure time-invariant filter.

% bss_decomp_gain - decompose an estimated source into target/interference/noise/artefacts components, assuming the admissible distortion is a pure time-invariant gain.

% bss_decomp_tvfilt - decompose an estimated source into target/interference/noise/artefacts components, assuming the admissible distortion is a time-varying filter.

% bss_decomp_tvgain - decompose an estimated source into target/interference/noise/artefacts components, assuming the admissible distortion is a time-varying gain.

% bss_make_frames - decompose some signal(s) into frames

% bss_make_lags - create a matrix containing lagged versions of some signal(s).

% bss_proj - compute the orthogonal projection of x on the subspace spanned by the row(s) of Y.

% bss_tvproj - compute the orthogonal projection of x onto the space of shifted windowed versions of the row(s) of Y

% bss_energy_ratios - compute energy ratios corresponding to SDR/SIR/SNR/SAR given a decomposition of an estimated source into target/interference/noise/artifacts over frames.

% bss_eval_test - run some basic tests to check the good behaviour of the toolbox. Look at this file to have an example of use of the toolbox

% decompose an estimated source into target/interference/noise/artefacts components, assuming the admissible distortion is a pure time-invariant filter.

% components, assuming the admissible distortion is a pure time-invariant