essentia 安装

Compiling Essentia from source

• Essentia depends on (at least) the following libraries:

• 
  

• FFTW: for the FFT implementation

• libavcodec/libavformat/libavutil/libavresample (from the FFmpeg project): for loading/saving any type of audio files (optional)

• libsamplerate: for resampling audio (optional)

• TagLib: for reading audio metadata tags (optional)

• LibYAML: for YAML files input/output (optional)

Installing dependencies on Linux

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.

In the case of Essentia 2.1, the required version of libtag1-dev is greater or equal to 1.9.1. It 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 libavcodec-dev, libavformat-dev, libavutil-dev and libavresample-dev is greater or equal to 10.1. 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 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

Installing dependencies on Mac OS X

Install a scientific python environment first:

1. install Command Line Tools for Xcode: https://github.com/mxcl/homebrew/wiki/Installation

2. install homebrew (package manager): http://brew.sh/

3. install prerequisites: brew install pkg-config gcc readline sqlite gdbm freetype libpng

4. install python: brew install python --framework

5. install ipython and numpy: pip install ipython numpy

6. install matplotlib: pip install matplotlib

7. when launching ipython, use:

1. ipython --pylab if you have matplotlib >= 1.3

2. 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

Then run:

brew install libyaml fftw ffmpeg libsamplerate libtag

Installing dependencies on Windows

Essentia does compile and run correctly on Windows, however there is no Visual Studio project readily available, so you will have to setup one yourself and compile the dependencies too. We will be working on Windows installer in the near future.

Additional dependencies for building documentation (python, all platforms)

To build the documentation you will also need the following dependencies (you might need to run this command with sudo):

pip install sphinx pyparsing sphinxcontrib-doxylink docutils

Other useful dependencies:

pip install pyyaml   # make sure to have libyaml installed first

You might need to install pip before, if you are on Linux:

sudo apt-get install python-pip

Compiling Essentia

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

• Use the keys:

• --with-python to enable python bindings, --with-examples to build examples based on the library, --with-vamp to build vamp plugin wrapper.

NOTE: you must always configure at least once before building!

The following will give you a list of options:

./waf --help

To compile everything you’ve configured:

./waf

To install the C++ library and the python bindings (if configured successfully; you might need to run this command with sudo):

./waf install

To run the C++ base unit tests (optional, only test basic library behavior):

./waf run_tests

To run the python unit tests (optional, include all unittests on algorithms, need python bindings installed first):

./waf run_python_tests

To generate the full documentation (optional, need python bindings installed first):

./waf doc

Documentation will be located in doc/sphinxdoc/_build/html/ folder.

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).

Using pre-trained high-level models in Essentia

The 2.0.1 version of Essentia includes a number of pre-trained classifier models for genres, moods and instrumentation. In order to use them you need to:

• Install Gaia2 library (supported on Linux/OSX): https://github.com/MTG/gaia/blob/master/README.md

• Build Essentia 2.0.1 with examples

• Use streaming_extractor_music (see detailed documentation)