openSMILE: The Munich Versatile and Fast Open-Source Audio Feature Extractor.
Authors: Florian Eyben, Felix Weninger, Martin Woellmer, Bjoern Schuller
NEW RELEASE APPROACHING (May 2014): A next major release (version 2.0) is on its way. Compared to 2.0-rc1 there have been many fixes and improvements and few new features. Expect the new final to be available in Mai or June this year, with all binary releases and updated documentation.
New release candidate (May-17 2013): We've released the first release candidate today (2.0-rc1). This is a source only release. Compilation is supported on Linux with autotools and g++/gcc, and with Visual Studio 2010 (and possibly later) on Microsoft Windows Vista or higher. The release candidate is available as a tar.gz file in the download area. Please note that the SVN here on sourceforge is not maintained anymore at the moment.
UPDATE (Dec-12 2012): Since we've got a few enquiries whether development was still active (due to the last release being two years old, and SVN activity has been halted), here is the latest news: Yes, the development is still very active. The main development has been moved to a private repository, from which we will publish releases from time to time. A next major release (version 2.0) is planned for February 2013, so please be patient. This will include a few new feature extractor components, performance and stability updates, bug fixes, as well as all recent feature sets for the INTERSPEECH and AVEC challenges.
The openSMILE feature extration tool enables you to extract large audio feature spaces in realtime. It combines features from Music Information Retrieval and Speech Processing. SMILE is an acronym for Speech & Music Interpretation by Large-space Extraction. It is written in C++ and is available as both a standalone commandline executable as well as a dynamic library. The main features of openSMILE are its capability of on-line incremental processing and its modularity. Feature extractor components can be freely interconnected to create new and custom features, all via a simple configuration file. New components can be added to openSMILE via an easy binary plugin interface and a comprehensive API.
A brief summary of openSMILE's features is given here (see the openSMILE book for a detailed and most up-to-date description of the features):
[ top ]
openSMILE is free software distributed under the GPL license up to version 1.0.1. Version 2.0 and above are distributed under a research only license. Customised commercial licensing options are available upon request, please contact info at audeering.com for details, or visit www.audeering.com (more information on the site coming soon).
[ top ]
We provide a release package which contains the linux and windows binaries as well as the source code and example configurations. A .zip file is provided for Windows systems and a .tar.gz file for Unix systems (the contents of both files are the same, however). The latest version is available for download from sourceforge here.
The latest version (only source package!) is version 2.0-rc1 (release candidate), released May 17th 2013.
The previous version is 1.0.1, released March 23rd 2010. This is a bugfix release.
The last 1.x series development code can be obtained from the Subversion repository. The repository is not maintained anymore, as we have moved development to a differnt repository due to various internal constraints. From version 2.0 on, please only use the released source packages. To get the old code from the SVN, type the following command in a command-line prompt on a system where SVN (http://subversion.tigris.org) is installed:
svn co https://opensmile.svn.sourceforge.net/svnroot/opensmile/trunk opensmile
openSMILE is well documented in the openSMILE book, which is available in electronic form as PDF. It is included with every release version, and can be also downloaded here: download openSMILE book. Installation instructions are also provided in the book. For the impatient: look for a suitable SMILExtract binary in the bin/ subdirectory of the release package and run it, run the script buildStandalone.sh to compile from source on Linux/Unix, or compile the Visual Studio solutions in ide/vs10 on Windows using Visual Studio 2010 or later. ... Then read then openSMILE book...
[ top ]
If you use openSMILE for your research, please cite the following paper for version 2.0rc1 and above:
Florian Eyben, Felix Weninger, Florian Gross, Björn Schuller: "Recent Developments in openSMILE, the Munich Open-Source Multimedia Feature Extractor", In Proc. ACM Multimedia (MM), Barcelona, Spain, ACM, ISBN 978-1-4503-2404-5, pp. 835-838, October 2013. doi:10.1145/2502081.2502224
For version 1.0.1 and below, you may cite the previous paper:
Florian Eyben, Martin Wöllmer, Björn Schuller: "openSMILE - The Munich Versatile and Fast Open-Source Audio Feature Extractor", In Proc. ACM Multimedia (MM), ACM, Florence, Italy, ACM, ISBN 978-1-60558-933-6, pp. 1459-1462, October 2010. doi:10.1145/1873951.1874246
We are always happy to hear what people are using openSMILE for. Thus, we would appreciate it, if you would send us a brief note with a reference to your paper, and/or a brief description of your work.
[ top ]
Additional help is provided by the authors and the community via the public forums on sourceforge.net. If you cannot find an answer to your problem there, please contact Florian Eyben via e-mail (eyben at tum . de).
[ top ]
Acknowledgment: openSMILE's development has received funding
from the European Community’s Seventh Framework
Programme (FP7/2007-2013) under grant agreement No.