SLEEPING-DISCO 9M: A large-scale pre-training dataset for generative music modeling

TL;DR

Sleeping-DISCO 9M is a large-scale, high-quality dataset of popular music and songs designed to advance generative music modeling tasks like text-music, captioning, and singing synthesis, reflecting real-world music.

Contribution

This paper introduces Sleeping-DISCO 9M, a novel open-source dataset of popular music from renowned artists, filling a gap in real-world music data for generative modeling.

Findings

Provides a high-quality dataset for various music generation tasks

Reflects real-world popular music and artist styles

Aims to improve adoption of datasets in the community

Abstract

We present Sleeping-DISCO 9M, a large-scale pre-training dataset for music and song. To the best of our knowledge, there are no open-source high-quality dataset representing popular and well-known songs for generative music modeling tasks such as text-music, music-captioning, singing-voice synthesis, melody reconstruction and cross-model retrieval. Past contributions focused on isolated and constrained factors whose core perspective was to create synthetic or re-recorded music corpus (e.g. GTSinger, M4Singer) and arbitrarily large-scale audio datasets (e.g. DISCO-10M and LAIONDISCO-12M) had been another focus for the community. Unfortunately, adoption of these datasets has been below substantial in the generative music community as these datasets fail to reflect real-world music and its flavour. Our dataset changes this narrative and provides a dataset that is constructed using actual popular music and world-renowned artists.