StemGMD: A Large-Scale Audio Dataset of Isolated Drum Stems for Deep Drums Demixing - part 1

<p>We introduce StemGMD, a new large-scale dataset of isolated drum stems that builds upon the extensive MIDI collection found in <a href="https://magenta.tensorflow.org/datasets/groove">Magenta's Groove MIDI Dataset (GMD)</a>.</p> <p>GMD is a 13.6-hour corpus of expressive drum performances executed by ten drummers on a Roland TD-11 electronic drum kit. It contains 1150 MIDI files along with the corresponding full-kit audio mixtures.</p> <p>As a first step in creating StemGMD, we mapped the 22 different MIDI pitches found in the original files onto nine canonical instruments through the reduction scheme proposed &nbsp;in J. Gillick, A. Roberts, J. Engel, D. Eck, and D. Bamman, "Learning to groove with inverse sequence transformations," in International Conference on Machine Learning (ICML), vol. 97, 2019, pp. 2269&ndash;2279.</p> <p>Each of the nine resulting MIDI channels was manually synthesized as a 16-bit/44.1 kHz stereo WAV file using ten realistic-sounding acoustic drum kits sourced from the <a href="https://support.apple.com/en-me/guide/logicpro/lgsi2fb2509e/mac">Logic Pro X sample libraries</a>, i.e., Bluebird, Brooklyn, Detroit Garage, East Bay, Heavy, Motown Revisited, Portland, Retro Rock, Roots, and SoCal.</p> <p>As a result, StemGMD contains 1224 hours of audio, which correspond to more than 136 hours of full-kit mixtures. Moreover, StemGMD also contains single hits for each of the drum pieces at ten different velocities ranging from 30 to 127.</p> <p>To the best of our knowledge, StemGMD is the largest publicly available dataset of drums to date. Moreover, it is the first collection of single-instrument clips from all nine pieces in a canonical drum kit, making it well-suited for training deep drums demixing models.</p> <p>&nbsp;</p> <p><strong>*** THIS IS PART 1 OF 2 ***</strong></p> <p><strong>Download part 2 here:</strong> <a href="../records/7882857">https://zenodo.org/records/7882857&nbsp;</a> (now available!)&nbsp;</p> <p>After downloading both parts, run <strong>unzip_StemGMD.sh</strong> to build the dataset from the split archive files.<br>Once unzipped, StemGMD will take just over <strong>1.13 TB </strong>of memory.</p> <p>&nbsp;</p> <p>The dataset is made available under a <a href="https://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution 4.0 International (CC BY 4.0) License</a>.</p> <p>____________________________</p> <p>We employed the dataset in our paper titled "Toward Deep Drum Source Separation," published in Pattern Recognition Letters.&nbsp;</p> <div> <div> <div> <div> <p>Please, cite this work as: A. I. Mezza, R. Giampiccolo, A. Bernardini, and A. Sarti, "Toward Deep Drum Source Separation," Pattern Recognition Letters, vol. 183, pp. 86-91, 2024, doi: 10.1016/j.patrec.2024.04.026.</p> <pre>@article{mezza2024, title = {Toward deep drum source separation}, author = {Alessandro Ilic Mezza and Riccardo Giampiccolo and Alberto Bernardini and Augusto Sarti}, journal = {Pattern Recognition Letters}, volume = {183}, pages = {86-91}, year = {2024}, issn = {0167-8655}, doi = {https://doi.org/10.1016/j.patrec.2024.04.026} }</pre> </div> </div> </div> </div>