Published February 28, 2024 | Version 1.0
Dataset Open

FMAK: A Dataset of Key and Mode Annotations for the Free Music Archive

Creators

Contributors

Description

We present a new expert-labeled dataset for the evaluation of key detection containing 340 hours (5489 songs) of song-level key and mode annotations, spread across 17 genres. 

For each song, we provide annotations for:

  • FMA track id
  • Spotify URI (when available)
  • Key and mode

All the audio is collected in and distributed by the FMA dataset by Michael Defferrard, Kirell Benzi, Pierre Vandergheynst, and Xavier Bresson.

  • The FMA metadata is made freely available for public use under a Creative Commons license
  • We do not hold the copyright on the audio and distribute it under the license chosen by the artist
  • The dataset is meant for research purposes

Files

000-019.zip

Files (39.3 GB)

Name Size Download all
md5:b86f6414820c1422b2c6cdf87be1ef3a
4.2 GB Preview Download
md5:a2da8377fdbc1d3a1f54dd60aa7b8f9b
4.6 GB Preview Download
md5:d70babe5f66bdf3e821c42a8b8aafb9b
3.3 GB Preview Download
md5:f53fcba704fce27e5c7f3ec2532dcb44
3.9 GB Preview Download
md5:1520f067d7caaf0813780ff69bc4ba85
3.6 GB Preview Download
md5:186643746fcb1f4722a28d3eb9c6b99c
3.4 GB Preview Download
md5:8cf882609fc2f301621c2e9f9da03214
5.0 GB Preview Download
md5:84f0f036e3778ffd97c10b591f803d06
4.8 GB Preview Download
md5:4a307f019d3354064814f05d1dffa1e2
4.8 GB Preview Download
md5:88d7dbcca82189ed75b7baa5aa132fc1
1.7 GB Preview Download
md5:d80a03bc8659edc60e335bd7f6bdf12a
273.3 kB Preview Download

Additional details