Benchmarking Music Demixing Models for Deep Drum Source Separation
Audio Samples
The following audio samples are provided as supplementary material for the article Benchmarking Music Demixing Models for Deep Drum Source Separation, authored by A. I. Mezza, R. Giampiccolo, A. Bernardini, and A. Sarti, and presented at the IEEE 5th International Symposium on the Internet of Sounds.
The first three examples are taken from the StemGMD dataset, which is composed of both mixtures and ground-truth stems. The last, instead, is taken from the musdb18hq dataset, which provides only mixtures, and thus no ground-truth tracks are present.