This is the accompanying Python toolbox for the Dagstuhl ChoirSet:
Sebastian Rosenzweig, Helena Cuesta, Christof Weiß, Frank Scherbaum, Emilia Gómez and Meinard Müller
Dagstuhl ChoirSet: A Multitrack Dataset for MIR Research on Choral Singing.
Transactions of the International Society for Music Information Retrieval (TISMIR), 3(1): 98–110, 2020.
- Download the DCS dataset from Zenodo
- Clone repository
- `git pull hidden URL
cd ./DCStoolbox
git submodule init
git submodule update
- Install Miniconda
conda env create -f environment.yml
source activate DCS
To use the toolbox functions type:
import DCStoolbox
If you want to contribute to this repository, please install nbstripout:
nbstripout --install
- demo1_dataParser.ipynb: Demo of dataset parser.
- demo2_F0Estimation.ipynb: F0-estimation using pYIN and CREPE.
- eval1_F0Annotations.ipynb: Evaluation of automatically extracted F0-trajectories on manual annotations.