First, make sure that all python dependencies are installed: conda env create

• cd ~/ && wget http://essentia.upf.edu/documentation/extractors/essentia-extractors-v2.1_beta2-linux-x86_64.tar.gz
• tar xzf essentia-extractors-v2.1_beta2-linux-x86_64.tar.gz
• echo 'PATH=~/essentia-extractors-v2.1_beta2/:$PATH' >> ~/.bashrc
• source ~/.bashrc

• Download tar for windows at http://essentia.upf.edu/documentation/extractors/
• Extract and add to the PATH variable

Place a file features.json in the input folder. It should contain settings about which files to analyze and which features to use for each file. Example:

{
  "flute.wav": [
    "spectral_energy",
    "spectral_centroid"
  ],
  "hihat.wav": [
    "spectral_energy"
  ]
}

Wav files referenced in the features.json file should reside in the same folder. They should have the following format:

• Number of channels: 1 (mono) or 2 (stereo)
• Sampling rate: 44100 Hz
• Bit depth: 16 or 32

Now run python analyzer.py and wait. It should create a file audioAnalysis.js in the input folder.