This dataset contains 30s excerpts from the Signal Separation Evaluation Campaign (SiSEC 2018). It accompanies the objective scores as submitted via the official github repository. The excerpts were generated using a method as described here that selects 30 seconds from each audio track by determining the most active parts of each source.

{"references": ["St\u00f6ter, Fabian-Robert; Liutkus, Antoine; Ito, Nobutaka, The 2018 Signal Separation Evaluation Campaign, arXiv:1804.06267"]}