Since the start of quarantine I've been live-streaming Catholic Masses where the choir and schola take pride in their ability to sing polyphony and chant which you typically never hear these days and which has always been in the public domain. However, because commercial recordings exist of these works, every single Sunday I've got to file protests against the copyright notices, pointing out that these are live, original performances of works that were in the public domain for over 200 years (over 1500 in some cases). No matter, if a recording of this music exists then CLEARLY no human could have possibly duplicated, in the year 2020, a melody and text that was already ancient a millennium ago.

I realize this is machine learning and music sampling but I'm a tad surprised that the dataset cannot include a field such as "this recording is of public domain music"... and after a YouTube channel has to keep saying "No, I'm not playing someone else's CDs, we have a choir singing these live -- can't you hear the imperfections and that the soprano missed that eighth?" I suspect that at some point the channel will be cancelled because its too much of a copyright hazard.