musiscape

A Python toolbox for analysing large music collections and long music recordings.

Point it at a folder of audio files, where each subfolder counts as an album, and it renders figures, tables and thumbnails that let you compare many tracks at a glance: albums, similarity, outliers, categories. Point it at a concert recording instead and it finds the songs inside it first, so the same tools apply to a live set.

Install

pip install musiscape

Quickstart

musiscape report ~/Music/my-collection

One command runs the whole pipeline and writes analysis/README.md next to your music: an album table, a similarity landscape, self-explaining categories, and the corpus extremes.

Commands

A live recording is not a collection until it is cut into songs, which is what segment is for (video files included, since concerts usually arrive as video):

musiscape segment ~/video/concert -o ~/video/concert/analysis
musiscape report  ~/video/concert/analysis/songs

Common options: -o sets the output folder (default <root>/analysis), --workers N parallelises, --duration S analyses only the first S seconds per track, and -k fixes the number of categories.

Learn more

• Documentation—guides, an illustrated gallery of the visualisation styles, and the API reference
• Wiki—methodology notes and a case study

The toolbox is meant for quick visualisations and overviews. Combine it with listening.

The four toolboxes

Four packages from the fourMs lab at the University of Oslo, each released separately on PyPI. Which one you want is decided by what you have in hand rather than by what you want to know:

Where a measure appears in more than one package it has a single owner and a single implementation, so the answer does not depend on which package you called. Circular statistics belong to micromotion, and musiscape imports pulse clarity, fifths-circle centres and the Rayleigh test from there rather than keeping its own copy. A test in each package checks its numbers against the owner's and fails if they diverge.

musiscape requires micromotion, which pip installs for you; that is the only sibling it needs. Its own boundary with ambiscape is worth stating, since both read audio: ambiscape is about a PLACE and its sonic ambience, musiscape about a COLLECTION of music. Music analysis moved out of ambiscape into musiscape on 2026-08-12, so a release of either from before then may still carry the other's functions.

Licence and credit

MIT licence. musiscape is developed as part of the AMBIENT project at RITMO Centre for Interdisciplinary Studies in Rhythm, Time and Motion, University of Oslo.

Citing

Cite the concept DOI, which always resolves to the newest version:

Jensenius, A. R. (2026). musiscape: A Python toolbox for analysing large music collections and long music recordings [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.21964192

Where the exact behaviour matters, add the version you ran. Every release has its own DOI, listed on the Zenodo record.

An older concept DOI, https://doi.org/10.5281/zenodo.21948999, is frozen at 0.5.0. It was created by a hand deposit made on 2026-08-15, before the Zenodo GitHub integration was archiving this repository; the integration began working the next day and every release since is under the DOI above. Zenodo cannot merge two concepts, so both records exist and only one of them advances. Cite the DOI above.

CITATION.cff in this repository carries the same information in machine-readable form.