This repository contains the corresponding code for the article:
Stefan Balke, Julian Reck, Christof Weiß, Jakob Abeßer, and Meinard Müller: JSD: A Dataset for Structure Analysis in Jazz Music, Transactions of the International Society for Music Information Retrieval, 2022.
The Jazz Structure Dataset (JSD) comprises structure annotations for 340 famous Jazz recordings. Along with the temporal annotations for song regions (e.g., Solo 1 starting at 40 s and ending at 113 s), it provides further metadata about the predominant instrument (in most cases the soloist) and the accompanying instruments (e.g., drums and piano).
This repository contains the annotations and reference implementations for the baselines described in the paper.
conda env create -f environment.yml
conda activate jsd
data/annotations_sv
: Original annotation files. Can be opened with SonicVisualiser.data/annotations_raw
: Direct export from SonicVisualiser into a textual format.data/annotations_csv
: Final annotations in CSV format.data/annotations_jams
: Final annotations in Jams format.If you want to build the annotations from ground up or integrate changes,
please call python prepare_annotations.py
This script takes the files from data/annotations_raw
and applies the following modifications:
data/track_durations.csv
).The usage of the baselines is explained in a separate Readme in the baselines
folder.
python general_statistics
data/general_statistics
.