Medley2K: A Dataset of Medley Transitions

TL;DR

Medley2K is a new dataset of 2,000 medleys with 7,712 labeled transitions across genres, designed to advance research in automatic medley generation and transition modeling.

Contribution

The paper introduces Medley2K, a comprehensive dataset for studying musical medley transitions, and demonstrates its utility by training a generative model for song transitions.

Findings

The dataset contains diverse transitions across genres.

A state-of-the-art model successfully generates song transitions.

The dataset facilitates research in automatic medley creation.

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.