Essentia: C++ library for audio analysis and music information retrieval, README overview

What it supplies: The README describes Essentia as an open-source C++ library for audio analysis and audio-based music information retrieval, released under the Affero GPLv3 license. It contains reusable algorithms for audio input/output, signal processing, statistics and a large set of spectral, temporal, tonal and high-level music descriptors. Useful for: - Compute music descriptors from audio using the library in C++ or Python for prototyping and research experiments. - Run predefined command-line extractors to compute common music descriptors without writing code. - Visualise results in Sonic Visualiser using the included Vamp plugin. First action: - On Linux x86_64 or i686, install the Python package with pip install essentia, or pip install essentia-tensorflow. - Or use the Docker images, or download prebuilt static binaries of some command-line extractors instead of installing the whole library. - Follow the Python quick start, the online tutorial or the Jupyter notebook, or use the command-line extractors. Prerequisites and limits: The README says Essentia is cross-platform on Linux, macOS, Windows, iOS and Android and names the pip route for Linux x86_64 and i686. It says to install from master for the latest updates, lists official releases on GitHub, and warns that behaviour can differ between versions. - Descriptor outputs can differ between Essentia versions, so record the version you use. - The README gives pip installation only for Linux x86_64 and i686; other platforms need the installation guide, which was not read. - The README does not list the descriptors or algorithms individually. The README says Essentia is released under the Affero GPLv3 license. That licence has network-use obligations beyond the ordinary GPL, so read the licence before using it in a service. The licence file was not read here. Audio you analyse carries its own rights. Written from the README text only. Statements are as the README gives them; nothing was installed or run and no audio was analysed. Based on README text only; the documentation site, the installation guide, the licence file and the code were not read. No install or run was made. Source inspected: https://raw.githubusercontent.com/MTG/essentia/23ae7baddecfc3227ba23ad69fa11f73ac5293a4/README.md ; 4251 bytes ; sha256 39f559194a4f6055a5daeb34699033216e86771ff55563b033953df932c9d420 ; 11 exact pinned excerpts in the draft pin map

{
  "revision": {
    "ref": "commit 23ae7baddecfc3227ba23ad69fa11f73ac5293a4",
    "pinned": false,
    "source_url": "https://raw.githubusercontent.com/MTG/essentia/23ae7baddecfc3227ba23ad69fa11f73ac5293a4/README.md"
  },
  "checked_at": "2026-11-07T10:10:08.988Z",
  "usefulness": "- Compute music descriptors from audio using the library in C++ or Python for prototyping and research experiments.\n- Run predefined command-line extractors to compute common music descriptors without writing code.\n- Visualise results in Sonic Visualiser using the included Vamp plugin.",
  "source_type": "repository",
  "availability": "available",
  "source_owner": "Music Technology Group, Universitat Pompeu Fabra (declared repository namespace)",
  "known_limitations": [
    "The README says Essentia is cross-platform on Linux, macOS, Windows, iOS and Android and names the pip route for Linux x86_64 and i686. It says to install from master for the latest updates, lists official releases on GitHub, and warns that behaviour can differ between versions.\n\n- Descriptor outputs can differ between Essentia versions, so record the version you use.\n- The README gives pip installation only for Linux x86_64 and i686; other platforms need the installation guide, which was not read.\n- The README does not list the descriptors or algorithms individually.\n\nThe README says Essentia is released under the Affero GPLv3 license. That licence has network-use obligations beyond the ordinary GPL, so read the licence before using it in a service. The licence file was not read here. Audio you analyse carries its own rights.\n\nWritten from the README text only. Statements are as the README gives them; nothing was installed or run and no audio was analysed.\nBased on README text only; the documentation site, the installation guide, the licence file and the code were not read. No install or run was made.",
    "Rendered README text of the exact archived capture at the pinned commit, read in full by the author; no media viewed or heard; nothing executed or installed.",
    "Source-documented steps; execution not performed. Internal affiliated review is not independent validation."
  ]
}