Tool that features audio extraction algorithms. openEAR is the Munich Open-Source Emotion and Affect Recognition Toolkit developed at the Technische Universit t M nchen (TUM). It provides efficient (audio) feature extraction algorithms implemented in C++, classfiers, and pre-trained models on well-k.

You can chose between binary releases for Linux and Windows, or get the source code and compile it yourself. The latter is the recommended way for Linux/Unix and MacOS systems. A binary release contains the main executable SMILExtract or SMILExtract.exe, example conguration les in the config/ folder, pre-trained models in the models/ folder, and scripts for visualisation and model-training in the scripts/ folder.

The binary releases are ready to use. For Linux, you must choose the executable you want to use. Executables for dierent architectures are provided in the linux bin directory. It should be obvious from the le name, which to use. Rename or copy the executable which suits your platform to SMILExtract in the top-level directory. Executables which are linked against PortAudio carry a pa in the le name.

For Windows, if you want to use PortAudio recording for live emotion recognition you must rename SMILExtract pa.exe to SMILExtract.exe and openSmileLib pa.dll to openSmileLib.dll. To test if your release works, change to the toplevel directory of the distribution and type ./SMILExtract -h

The license of this software is Free, you can free download and free use this calculator software.