
A dataset suite designed for music, SFX, and spatial audio
Commercially usable audio datasets structured around the primitives models need to learn: stems, vocals, MIDI, mix pairs, Foley, SFX, spatial scenes, room data, human preference labels, and more.
