PDMX: A Large-Scale Public Domain MusicXML Dataset for Symbolic Music Processing

We introduce PDMX: a Public Domain MusicXML dataset for symbolic music processing. Refer to our paper for more information, and our GitHub repository for any code-related details. Please cite our paper if you use this dataset. Upon further use of the PDMX dataset, we discovered a discrepancy between the public-facing copyright metadata on the MuseScore website and the internal copyright data of the MuseScore files themselves, which affected 31,221 (12.29% of) songs. We have decided to proceed with the former given its public visibility on Musescore (i.e. this is what the MuseScore website presents its users with). We have noted files with conflicting internal licenses in the license_conflict column of PDMX. We recommend using the no_license_conflict subset of PDMX (which still includes 222,856 songs) moving forward. Additionally, for each song in PDMX, we not only provide the MusicRender and metadata JSON files, but we also try to include the associated compressed MusicXML (MXL) and sheet music (PDF) files when available. Due to the corruption of 42 of the original MuseScore files, these songs lack associated MXL and PDF files (since they could not be converted to these formats) and only include the MusicRender and metadata JSON files. The valid_mxl_pdf subset of PDMX describes all songs with valid associated MXL and PDF files.