The SBM Matlab toolbox for “Supervised Binaural Mapping”, contains a set of functions and scripts for supervised binaural sound source separation and localization. The approach consists in learning the acoustic space of a system using a set of white-noise measurements. Once the acoustic space is learned, it can be used to efficiently localize one or several natural sound sources such as speech, and to separate their signals.
This toolbox include:
- The algorithm Probabilsitic Piecewise Affine Mapping (PPAM) published in 
- The algorithm Variational EM for Sound Source Separation and Localization (VESSL) published in [2,3]
- 7 usage examples demonstrating single and multiple sound source localization, sound source separation, and online audio-to-video mapping
This toolbox requires:
- The AVASM dataset : https://team.inria.fr/robotlearn/the-avasm-dataset/
- The GLLiM Matlab toolbox: https://team.inria.fr/robotlearn/gllim_toolbox/
- Optionally, the BSS Eval Matlab toolbox V3.0: http://bass-db.gforge.inria.fr/bss_eval/
This video is one example of the results which can be generated from the toolbox:
To learn more…
For more information, we invite you to visit our related research pages:
For any questions, comments or feedback, please contact Antoine Deleforge.
 A. Deleforge, R. Horaud, Y. Schechner, L. Girin, “Co-Localization of Audio Sources in Images Using Binaural Features and Locally-Linear Regression”, IEEE/ACM Transactions in Audio, Speech and Language Processing, 2015.
 A. Deleforge, F. Forbes, R. Horaud, “Acoustic space learning for sound-source separation and localization on binaural manifolds” International Journal of Neural Systems, 2015.
 A. Deleforge, F. Forbes, R. Horaud, “Variational EM for Binaural Sound-Source Separation and Localization“, IEEE International Conference on Acoustic, Speech, and Signal Processing, Vancouver, CA, 2013.