** Apologies for any cross-posting **
Dear list,
We are pleased to announce the first release of MedleyDB, a dataset of multitrack audio with melody annotations and instrument activation annotations:
MedleyDB contains 122 multitracks (mix + stems + raw audio) from diverse genres. It is released to the research community under a Non-commercial Creative Commons license. The dataset was curated to facilitate research in topics such as melody extraction, instrument recognition, source separation and automatic mixing, among others. For more information, please visit MedleyDB’s companion website.
MedleyDB will be presented at ISMIR next week. For those interested, please see our paper:
R. M. Bittner, J. Salamon, M. Tierney, M. Mauch, C. Cannam and J. P. Bello. "MedleyDB: A Multitrack Dataset for Annotation-Intensive MIR Research", in Proc. 15th International Society for Music Information Retrieval Conference (ISMIR 2014), Taipei, Taiwan, October 2014.
We hope you find the dataset useful for your work and look forward to seeing some of you at ISMIR.
Best,
Rachel, Justin, Mike, Matthias, Chris and Juan.