DEAM dataset consists of 1802 excerpts and full songs annotated with valence and arousal values both continuously (per-second) and over the whole song. The detailed description of the dataset is given in the Manual. The metadata describing the audio excerpts (their duration, genre, folksonomy tags) is in the metadata archive.
Download the files
Both the annotations, the Creative Commons licensed sound files, and the features extracted with openSMILE are available:
- Download the audio (1.3 Gb)
- Download the annotations (4.7 Mb)
- Download the features (601.5 Mb)
For more details please consult the manual.