English

Medley2K: A Dataset of Medley Transitions

Sound 2020-08-26 v1 Machine Learning Audio and Speech Processing

Abstract

The automatic generation of medleys, i.e., musical pieces formed by different songs concatenated via smooth transitions, is not well studied in the current literature. To facilitate research on this topic, we make available a dataset called Medley2K that consists of 2,000 medleys and 7,712 labeled transitions. Our dataset features a rich variety of song transitions across different music genres. We provide a detailed description of this dataset and validate it by training a state-of-the-art generative model in the task of generating transitions between songs.

Keywords

Cite

@article{arxiv.2008.11159,
  title  = {Medley2K: A Dataset of Medley Transitions},
  author = {Lukas Faber and Sandro Luck and Damian Pascual and Andreas Roth and Gino Brunner and Roger Wattenhofer},
  journal= {arXiv preprint arXiv:2008.11159},
  year   = {2020}
}

Comments

MML 2020 - 13th Int. Workshop on Machine Learning and Music at ECML-PKDD 2020