MTG/essentia

What it solves

Essentia provides a comprehensive toolkit for analyzing audio and retrieving music information. It solves the problem of needing a robust, computationally efficient way to extract spectral, temporal, and tonal descriptors from audio files, which is essential for music research and industrial audio applications.

How it works

It is a C++ library wrapped in Python that implements a wide range of digital signal processing (DSP) blocks and algorithms. It offers predefined executable extractors for fast prototyping and includes a Vamp plugin for visualization via Sonic Visualiser. The library is designed to be cross-platform, supporting Linux, macOS, Windows, iOS, and Android.

Who it’s for

This tool is designed for researchers conducting music information retrieval experiments and developers building large-scale industrial audio applications.

Highlights

• Extensive collection of reusable algorithms for audio I/O and signal processing.
• Optimized for computational cost and robustness of music descriptors.
• Python wrappers for rapid prototyping.
• Cross-platform support including mobile operating systems.