metadata

license: mit

Introduction

This is a data subset extracted from the FMA music dataset, particularly the fma_small subset. There are 7 top level genres in this dataset, namely:

Hip-hop
Pop
Folk
Rock
International
Electronic
Instrumental

The dataset is composed of 210 .mp3 tracks, each of 30 seconds.

Metadata of the audio mp3 audio tracks could be found in metadata.csv.

The full mfa dataset could be found here

Credits

All credits go to the original creators for this dataset

@inproceedings{fma_dataset,
  title = {{FMA}: A Dataset for Music Analysis},
  author = {Defferrard, Micha\"el and Benzi, Kirell and Vandergheynst, Pierre and Bresson, Xavier},
  booktitle = {18th International Society for Music Information Retrieval Conference (ISMIR)},
  year = {2017},
  archiveprefix = {arXiv},
  eprint = {1612.01840},
  url = {https://arxiv.org/abs/1612.01840},
}