You can install those dependencies on a Debian/Ubuntu system from official repositories using the commands provided below. Note that, depending on the version of Essentia, different versions of libav* and libtag1-dev packages are required. See Github release notes on the download page.
In the case of Essentia 2.1, the required version of TagLib (libtag1-dev) is greater or equal to 1.9. The suitable version is distributed with Ubuntu Trusty (14.04 LTS). If you are using the latest stable Debian (Wheezy), you might want to install it from wheezy-backports repository. The required version of LibAv (libavcodec-dev, libavformat-dev, libavutil-dev and libavresample-dev) is greater or equal to 10. The appropriate versions are distributed in Ubuntu Utopic (14.10) repository, and in Debian wheezy-backports.
Essentia 2.1 on Ubuntu 14.10:
sudo apt-get install build-essential libyaml-dev libfftw3-dev libavcodec-dev libavformat-dev libavutil-dev libavresample-dev python-dev libsamplerate0-dev libtag1-dev
In order to use python bindings for the library, you might also need to install python-numpy-dev or python-numpy on Ubuntu:
sudo apt-get install python-numpy-dev python-numpy
Install support for YAML files input/output in python (optional, make sure to have libyaml installed first):
sudo apt-get install python-pip pip install pyyaml
Install a scientific python environment first:
- ipython --pylab if you have matplotlib >= 1.3
- ipython --pylab=tk if you have matplotlib < 1.3
Note that you are advised to install python environment as described here, i.e., via homebrew and pip. You will most probably encounter installation errors when using python/numpy preinstalled with OSX 10.9. More details can be found at https://github.com/mxcl/homebrew/wiki/Homebrew-and-Python
brew install libyaml fftw ffmpeg libsamplerate libtag
Once your dependencies are installed, you can compile Essentia (the library) by going into its directory and start by configuring it:
./waf configure --mode=release --with-python --with-cpptests --with-examples --with-vamp --with-gaia
NOTE: you must always configure at least once before building!
The following will give you a list of options:
To compile everything you’ve configured:
To install the C++ library and the python bindings (if configured successfully; you might need to run this command with sudo):
All built examples (including the out-of-box features extractors) will be located in build/src/examples/ folder, as well as the vamp plugin file libvamp_essentia.so. In order to use the plugin you will need to place this file to the the standard vamp plugin folder of your system (such as /usr/local/lib/vamp/ on Linux).
If you want to assure that Essentia works correctly, do the tests.
To run the C++ base unit tests (only test basic library behavior):
To run the python unit tests (include all unittests on algorithms, need python bindings installed first):
All documentation is provided on the official website of Essentia library. To generate it by your own follow the steps below.
Install pip, if you are on Linux:
sudo apt-get install python-pip
Install additiona dependencies (you might need to run this command with sudo):
pip install sphinx pyparsing sphinxcontrib-doxylink docutils
Make sure to install Essentia with python bindings and run:
Documentation will be located in doc/sphinxdoc/_build/html/ folder.
Essentia does compile and run correctly on Windows (python bindings were not tested). The easiest way to build Essentia is by cross-compilation on Linux using MinGW: https://github.com/MTG/essentia/blob/master/FAQ.md#cross-compiling-for-windows-on-linux
However, if you want to use Visual Studio, there is no project readily available, so you will have to setup one yourself and compile the dependencies too. It appears that binaries for the library generated by cross-compilation are not compatible with Visual Studio.
Essentia includes a number of pre-trained classifier models for genres, moods and instrumentation. In order to use them you need to:
You can also use classifier models trained by your own: https://github.com/MTG/essentia/blob/master/FAQ.md#training-and-running-classifier-models-in-gaia