scikit‐maad: An open‐source and modular toolbox for quantitative soundscape analysis in Python

TLDR

Passive acoustic monitoring is increasingly applied to terrestrial, marine and freshwater environments, providing cost‑efficient biodiversity surveys, but processing the avalanche of audio recordings remains a major bottleneck that hinders research and conservation, and this development will create synergies among ecoacousticians to explore acoustic diversity. We present scikit‑maad, an open‑source Python package dedicated to the analysis of environmental audio recordings. The package loads and processes digital audio, segments regions of interest, computes acoustic features and sound‑pressure levels, provides field recordings and extensive documentation, and is released under a BSD license to support reproducible research. scikit‑maad enables efficient scanning of large audio datasets and easy integration of machine‑learning Python packages, allowing measurement of acoustic properties and identification of key patterns in all kinds of soundscapes.

Abstract

Abstract Passive acoustic monitoring is increasingly being applied to terrestrial, marine and freshwater environments, providing cost‐efficient methods for surveying biodiversity. However, processing the avalanche of audio recordings remains challenging, and represents nowadays a major bottleneck that slows down its application in research and conservation. We present scikit‐maad, an open‐source Python package dedicated to the analysis of environmental audio recordings. This package was designed to (a) load and process digital audio, (b) segment and find regions of interest, (c) compute acoustic features and (d) estimate sound pressure levels. The package also provides field recordings and a comprehensive online documentation that includes practical examples with step‐by‐step instructions for beginners and advanced users. scikit‐maad opens the possibility to efficiently scan large audio datasets and easily integrate additional machine learning Python packages into the analysis, allowing to measure acoustic properties and identify key patterns in all kinds of soundscapes. To support reproducible research, the package is released under the BSD open‐source licence, which allows unrestricted redistribution for commercial and private use. This development will create synergies between the community of ecoacousticians, such as engineers, data scientists, ecologists, biologists and conservation practitioners, to explore and understand the processes underlying the acoustic diversity of ecological systems.