MAD-EEG: an EEG dataset for decoding auditory attention to a target instrument in polyphonic music
By G.Cantisani, S. Essid, G. Richard

MAD-EEG is a research corpus for studying the problem of EEG-based auditory attention decoding for music stimuli. It contains EEG responses to music recorded while the subjects were attending to a particular instrument in a music mixture. 

The dataset is available from Zenodo. For further details, please refer to the paper: (cantisani et al., 2019).