Sound Field Synthesis Toolbox for Matlab

The SFS Toolbox for Matlab gives you the possibility to play around with sound field synthesis methods like wave field synthesis (WFS) or near-field compensated higher order Ambisonics (NFC-HOA). There are functions to simulate monochromatic sound fields for different secondary source (loudspeaker) setups, time snapshots of full band impulses emitted by the secondary source distributions, or even generate binaural room scanning (BRS) impulse response sets in order to generate binaural simulations of the synthesized sound fields with the SoundScape Renderer.

Source code and issue tracker:
SFS Toolbox for Python:
MIT – see the file LICENSE for details.


Download the Toolbox, go to the main path of the Toolbox and start it with SFS_start which will add all needed paths to Matlab/Octave. If you want to remove them again, run SFS_stop.


You need Matlab version R2011b or newer to run the Toolbox. On older versions the Toolbox should also work, but you need to add narginchk.m to the SFS_helper directory.
You need Octave version 3.6 or newer to run the Toolbox. In addition, you will need the audio and signal packages from octave-forge.
If audioread() is not available in your Matlab or Octave version, you can replace it by wavread(). It is used in the two functions auralize_ir() and compensate_headphone().
Impulse responses:
The Toolbox uses the SOFA file format for handling impulse response data sets like HRTFs. If you want to use this functionality you also have to install the SOFA API for Matlab/Octave, which you can add to your paths by executing SOFAstart.
Backward compatibility:
Since version 2.0.0 the SFS Toolbox incorporates SOFA as file format for HRTFs which replaces the old irs file format formerly used by the Toolbox. If you still need this you should download the latest version with irs file support.

Getting started

In order to make a simulation of the sound field of a monochromatic point source with a frequency of 800 Hz placed at (0,2.5,0) m synthesized by WFS run

conf = SFS_config;
conf.plot.normalisation = 'center';
sound_field_mono_wfs([-2 2],[-2 2],0,[0 2.5 0],'ps',800,conf)

To make a simulation of the same point source - now producing a broadband impulse - in the time domain at a time of 200 samples after the first loudspeaker activity run

conf = SFS_config;
conf.plot.normalisation = 'max';
sound_field_imp_wfs([-2 2],[-2 2],0,[0 2.5 0],'ps',200,conf)

After that have a look at SFS_config.m for the default settings of the Toolbox. Please don’t change the settings directly in SFS_config.m, but create an extra function or script for this, that can look like this:

conf = SFS_config;
conf.fs = 48000;

For a detailed description of all available features the SFS Toolbox, have a look at the online documentation.

Credits and feedback

If you have questions, bug reports or feature requests, please use the Issue Section to report them.

If you use the SFS Toolbox for your publications please cite our AES Convention e-Brief and the DOI for the used Toolbox version, you will find at the official releases page:

H. Wierstorf, S. Spors - Sound Field Synthesis Toolbox. In the Proceedings of 132nd Convention of the Audio Engineering Society, 2012 [ pdf ] [ bibtex ]

Copyright (c) 2010-2016 SFS Toolbox Developers