Sound Field Synthesis Toolbox 0.2 released

May 9, 2012 in Anouncements, MATLAB by Hagen Wierstorf | No comments

We are happy to announce the first public and open source release of our Sound Field Synthesis (SFS) Toolbox. It is a toolbox for Matlab/Octave to numerical simulate different SFS methods like Wave Field Synthesis (WFS) or higher order Ambisonics (HOA). It provides a variety of functions to reach mono-frequent and spatio-temporal impulse responses for different array geometries and driving functions.

You can find the sources and documentation on the corresponding SFS Toolbox site.

Note that there are two bugs in the current version, and one of them will prevent you from plotting the results. Have a look at the corresponding wiki page how to fix this.

To demonstrate some of the features of the toolbox, below an image of a mono-frequent plane wave synthesized using WFS and a linear loudspeaker array is shown. The different colors of the speakers at the ends of the array visualize the applied tapering window.

plane wave, synthesized using WFS and a linear loudspeaker array

Tags: , , , ,

Listening examples for the paper “Evaluation of perceptual properties of phase-mode beamforming in the context of data-based binaural synthesis” presented at ISCCSP 2012

May 2, 2012 in Papers by Sascha Spors | No comments

We have presented a paper at the 5th International Symposium on Communications, Control, and Signal Processing. It discusses the influence of spatial bandlimitation in modal beamforming when used as plane wave decomposition for binaural synthesis. Besides the instrumental measures given in the paper/presentation you can also download a series of listening examples. The samples have been computed for the parameters described in the paper. The maximum order N of the spherical harmonics expansion has been varied, as well as the head rotation (incidence angle of the virtual source). As source material noise bursts, speech and castanets have been used. The examples are provided as binaural sound files optimized for the AKG K-601 headphone. The file names encode the conditions and incidence angle.

Download the listening examples by following this link

 

The full reference to the paper is

S. Spors and H. Wierstorf. Evaluation of perceptual properties of phase-mode beamforming in the context of data-based binaural synthesis. In IEEE-EURASIP International Symposium on Control, Communications, and Signal Processing, May 2012.

Tags: , , , , ,

Listening examples for the paper “A comparison of phase-mode versus delay-and-sum beamforming in the context of data-based binaural synthesis”

April 28, 2012 in Papers by Sascha Spors | No comments

At the 132nd Convention of the Audio Engineering Society we presented the paper

S. Spors, H. Wierstorf, and M. Geier. Comparison of modal versus delay-and-sum beamforming in the context of data-based binaural synthesis. In 132th AES Convention. Audio Engineering Society (AES), April 2012.

The presentation slides are available here.

This post provides listening examples for the comparison of modal (MB) and delay-and-sum (DSB) beamforming when used as plane wave decomposition for binaural synthesis. The stimuli have been computed as described in the paper for spatially continuous and sampled spherical microphone arrays. Additionally for modal beamforming the maximum order of the spherical harmonics has been varied in the spatially continuous case. As source material noise bursts, speech and castanets have been used. The examples are provided for various incidence angles, as stereo sound files optimized for the AKG K-601 headphone. The file names encode the conditions and incidence angle.

Download the samples for

by following the links.

Tags: , , , , ,

First Release of the Audio Processing Framework (APF)

April 10, 2012 in Anouncements, Papers by Matthias Geier | No comments

We collected some generic C++ code which we programmed over the last few years and bundled it into the so-called Audio Processing Framework (APF).

That’s the website: http://tu-berlin.de/?apf

More details (including downloads and Doxygen documentation) are here: http://dev.qu.tu-berlin.de/projects/apf/wiki

It will be presented at the Linux Audio Conference (LAC) in a few days.

Here’s the paper, feel free to cite:

Matthias Geier, Torben Hohn and Sascha Spors: “An Open Source C++ Framework for Multithreaded Realtime Multichannel Audio Applications”, Linux Audio Conference, Stanford, CA, USA, April 2012.

Release 0.3.3 of the SoundScape Renderer

February 7, 2012 in Anouncements by Sascha Spors | No comments

We are proud to announce a new release of the SoundScape Renderer (SSR). The new release 0.3.3 (Harpsichord) can be downloaded here. This release contains a number of bug-fixes and extensions. Major improvements are

  • fixed a bug in the WFS renderer which caused audible artefacts (klicks)
  • support for our Razor-based head-tracker
  • support for OS X 10.7

We would like to thank all contributors who reported bugs or even provided bug-fixes. We encurage all users to send bugs, usage reports and feature requests to SoundScapeRenderer@telekom.de. You can also send us your (example) scenes or content you have produced with the SSR. Some user content can already be found here.

The SSR development team.

Tags:

New Release of SOFiA available

December 22, 2011 in Anouncements, MATLAB by Sascha Spors | No comments

A new release of the Sound Field Analysis Toolbox (SOFiA) is available! This release includes a number of bugfixes and extensions:

  • various bug-fixes and optimizations
  • new sampled wave generator that handels various array configurations (open/dual-open/rigid spheres, pressure/pressure-gradient microphones) for a plane wave/point source as incident sound field
  • native support for Legendre and Lebedev grids

SOFiA can be downloaded here http://code.google.com/p/sofia-toolbox/.

Tags: , ,

Book on Sound Field Synthesis

December 5, 2011 in MATLAB, Papers by Jens Ahrens | 1 comment

I proudly announce that some time ago, I submitted the manuscript of a monography on sound field synthesis to Springer who will publish it under the title Analytic Methods of Sound Field Synthesis in spring 2012. You can find the official announcement here.
The book is not only a mere copy-and-paste of our papers. All our own work has been significantly enhanced in didactical terms and, of course, all relevant works of other authors and a considerable amount of previously unpublished material are included. Besides the physical fundamentals and spatial discretization (e.g., spatial aliasing), extensive discussions of various applications of sound field synthesis such as moving virtual sound sources, focused virtual sound sources, etc. and a variety of practical aspects are discussed. I have also put some effort in clarifying the relationship between Ambisonics and Wave Field Synthesis.

MATLAB/Octave scripts for all simulations contained in the book will be provided at http://www.soundfieldsynthesis.org.

Tags: , , , , , ,

DIY low-cost head-tracker

November 30, 2011 in Anouncements, Videos by Sascha Spors | No comments

We have published a tutorial and firmware which allows to put together a low-cost Attitude and Heading Reference System (AHRS) that can be used as a head-tracker for virtual acoustics. A wired version of the tracker (connected via USB), as well as a wireless version (using Bluetooth) can be realized quite simple using comercially available hardware. The tracker is based on the Arduino-compatible “9DOF Razor IMU” board by SparkFun, which contains a 3-axis gyroscope/accelerometer/magnetometer and a microcontroller to do the sensor fusion.

Open source AHRS / head-tracking using SparkFun 9DOF Razor IMU (via USB or Bluetooth) from Spatial Audio Research on Vimeo.

The next release of the SoundScape Renderer (SSR) will support the tracker.

Tags: , , ,

SOFiA Sound Field Analysis Toolbox

November 13, 2011 in Anouncements, MATLAB by Sascha Spors | No comments

Spherical microphone arrays and sound field decomposition using spherical harmonics is a well established technique for the analysis of sound fields, room acoustics or spatial audio recording applications. The Sound Field Analysis Toolbox (SOFiA) provides a MATLAB toolbox for the analysis of sound fields using data from spherical microphone arrays.  It is freely available under the GNU GPL v3 license. For the evaluation of SOFiA, exemplary datasets from a VariSphear scanning array system are available under the Creative Commons license.

SOFiA can be downloaded here http://code.google.com/p/sofia-toolbox/

An overview of the SOFiA processing chain as well as a closer discussion of selected details can be found in the paper

B. Bernschütz, C. Pörschmann, S. Spors, and S. Weinzierl. SOFiA – Sound field analysis toolbox. In International Conference on Spatial Audio, Detmold, Germany, November 2011. [PDF]

Tags: , , ,

Using the RME MADIface on Linux

November 5, 2011 in Uncategorized by Matthias Geier | 1 comment

If you want to use the RME MADIface on Linux, it probably doesn’t work right away.

The latest version of ALSA (1.0.24) – which is probably already installed on your computer – is from January 2011. However, there was a lot of activity and bug-fixing going on since February. So you won’t have all the good stuff and a few things won’t work at all.

Especially, we couldn’t use the sound card in slave mode, and even in master mode a few sample rates just didn’t work. And, what’s really sad, the hdspmixer didn’t work.

But fear not, there is still hope!

That’s what we did:

Download the newest and freshest alsa-driver package from here: http://www.alsa-project.org/snapshot/.

Extract the package somewhere, and compile the thing:

tar xvjf /path/to/alsa-driver-1.0.24.88.gb41c8.1877.ga1e0a.tar.bz2
cd alsa-driver-1.0.24.88.gb41c8.1877.ga1e0a
./gitcompile

You may need a few development packages from your Debian (or whatever you are using) repository (maybe build-essentials, autoconf, …).

And then, as root, install everything:

make install

And that’s it! After re-booting (or restarting the relevant services if you know what those are …) everything should work. You can set the sync-options with alsamixer and you can also use the hdspmixer.

Of course, all this will be obsolete once the new ALSA version 1.0.25 (or whatever comes after 1.0.24) will be released as Debian (or whatever you are using) package.

Tags: ,

« Older entries