What is Essentia?
Essentia is an open-source C++ library for audio analysis and audio-based music information retrieval. It
provides a reusable collection of algorithms mainly used to extract features from audio files. The library provide
Python bindings, which means that you can control Essentia and write scripts that use Essentia’s features from
within Python (just as you would in Matlab). There is also a streaming mode in which you can connect algorithms and
run them automatically instead of specifying explicitly the order of execution (such as PureData or Max/MSP). The
advantage of the streaming mode is that it requires less boilerplate code and uses less memory.
Essentia is not a framework, but rather a collection of algorithms wrapped in a library. It doesn’t
provide common high-level logic for descriptor
computation (therefore, you are not locked into a certain way of doing things). It rather focuses on the robustness,
performance and optimality of the provided algorithms, as well as ease of use. The flow of the analysis is decided
and implemented by the user, while Essentia is taking care of the implementation details of the algorithms being
used. A number of predefined executable extractors for the available music descriptors are provided with the library
as examples, however they should not be considered as the only correct way of doing things.
Essentia currently features the following algorithms (among others):
- Audio file input/output: ability to read and write nearly all audio file formats (wav, mp3, ogg, flac, etc.)
- Standard signal processing blocks: FFT, DCT, frame cutter, windowing, envelope, smoothing
- Filters (FIR & IIR): low/high/band pass, band reject, DC removal, equal loudness
- Statistical descriptors: median, mean, variance, power means, raw and central moments, spread, kurtosis, skewness, flatness
- Time-domain descriptors: duration, loudness, LARM, Leq, Vickers' loudness, zero-crossing-rate, log attack time and other signal envelope descriptors
- Spectral descriptors: Bark/Mel/ERB bands, MFCC, GFCC, LPC, spectral peaks, complexity, rolloff, contrast, HFC, inharmonicity and dissonance
- Tonal descriptors Pitch salience function, predominant melody and pitch, HPCP (chroma) related features, chords, key and scale, tuning frequency
- Rhythm descriptors beat detection, BPM, onset detection, rhythm transform, beat loudness
- Other high-level descriptors: danceability, dynamic complexity, audio segmentation, SVM classifier
|
Introduction
An introduction to Essentia's main concepts
Building and installing Essentia
Instructions to get Essentia running on your computer
Algorithms overview
A quick description of the main algorithms
Python tutorial for beginners
A hands-on introduction to Essentia
Using extractors out-of-box
Getting some quick results with no programming required
|
Design Overview
An explanation of Essentia's basic types and classes
"Standard" mode how-to
Learn how to write a "standard" extractor
Streaming mode architecture
A description of how the "streaming" mode works
"Streaming" mode how-to
Learn how to write a "streaming" extractor
Algorithm reference
The detailed documentation for all the algorithms
|
|
Extending Essentia
How to write new "standard" algorithms for Essentia
Extending Essentia "streaming"
How to write new "streaming" algorithms for Essentia
Composite API
The inner workings of the composite algorithms
Execution network algorithm
How Essentia's streaming scheduler works
|
Doxygen C++ documentation
The documentation for the base classes used in Essentia
Coding guidelines
Good practices to follow when developing new algorithms
Contents
For a complete overview of the documentation
|
{% endblock %}