ENSA is probably one of the most important results of my doctoral thesis. This dataset has been specially designed to study and design music recommender systems. It contains 60 audio files of original songs by Colombian artists. Additionally, different music appreciation experiments are also available in this repository, as well as emotional labeling by artists, machine learning models for emotion recognition in music with affective dimensional models, emotional evaluation of listeners, and other data. For a better understanding I recommend reading the scientific paper that supports the research.

https://link.springer.com/article/10.1007/s00779-023-01721-4 Ospitia-Medina, Y., Beltrán, J.R. & Baldassarri, S. ENSA dataset: a dataset of songs by non-superstar artists tested with an emotional analysis based on time-series. Pers Ubiquit Comput 27, 1909–1925 (2023). https://doi.org/10.1007/s00779-023-01721-4

For more information you can write to me at yesid.ospitia@gmail.com