Many of the recent controversies around creative visual AIs – those that create images mainly – have focused on training materials. What if the problem for musical AIs is not that they have been trained on copyrighted music, but that they haven’t? “The thing with AI is that the output is only as good as the input. You need high-quality data sets to train your models on,” Oleg Stavitsky, CEO of AI music startup Endel, said. “Most of the AI music models were trained on just stock music, or stems that were created by a bunch of session musicians basically.”
Source: Music’s problem with AI training may not be the one you expect – Music Ally