Stable-audio-demo shows where we're at in AI-generated music. The broad realm of production/library music, the sort of intentionally unemotive instrumental noodling you hear in corporate training videos and local advertisements, is surely a goner. The question is whether (and how fast) it will be good enough to market to consumers. Lyrics and human voices are not attempted; an example with whistling is uncanny, like an invisible wet tentacle finding your ankle.
Prompt: This song contains someone strumming a melody on a mandolin while more people are whistling along. Then a mandolin, an e-bass and an acoustic guitar are playing a short melody in a lower key before breaking into the next part along with flutes and percussions. This song may be played outside by musicians performing.
Note how it's obviously superior to the older model it's compared against, but to my mind has rather less character (Brian Eno: "Whatever you now find weird, ugly, uncomfortable and nasty about a new medium will surely become its signature.")
That said, i'm a big fan of Disco, Driving Drum Machine, Synthesizer, Bass, Piano, Guitars, Instrumental, Clubby, Euphoric, Chicago, New York, 115 BPM. Best jam in hours.
P.S. I wonder if a killer app for this might be reprocessing stems, a second stage to restore quality lost in the extraction once an instrument has been isolated.
