Freesound Datasets is a Zenodo community to host datasets based on Freesound audio content.

Some datasets are collected through the Freesound Annotator, a platform for the collaborative creation of open audio collections labeled by humans and based on Freesound content. Freesound Annotator is created and maintained by the Music Technology Group (MTG) of the Universitat Pompeu Fabra in Barcelona.

Other datasets have been annotated through other means, often developed also at the Music Technology Group.

https://annotator.freesound.org/

https://www.upf.edu/web/mtg