AI and the Weird Future of Music

Rolling Stone Music Now

Rolling Stone

Music Commentary, Music, Music Interviews

4 • 1K Ratings

🗓️ 1 May 2023

⏱️ 36 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

Is artificial intelligence about to change music forever? We look at what's already happening — and what's coming next. Grimes, for one, is allowing anyone to use AI models of her voice, and splitting royalties 50/50; her manager, Daouda Leonard, joins host Brian Hiatt to talk about their plans and the broader risks and possibilities of AI in music. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Transcript

Click on a timestamp to play from that location

0:00.0	Hey, I'm Brian Hyatt, and this is Rolling Stone Music Now.
0:05.0	2023 appears to be the year of artificial intelligence.
0:09.0	Chat TBT is of course sparking a lot of discussion and fear about the future of white-collar work and school assignments and various other things.
0:18.0	And meanwhile, the music business is starting to have some of the same conversations, not really because of Chad TBT.
0:23.4	A couple weeks ago a song called Heart on my sleeve by someone who calls themselves Ghost
0:28.2	Rider went up briefly on streaming services.
0:31.0	Ghost Rider used AI voice cloning software to make it sound like Drake in the weekend we're singing this song. Bumping Justin Gossin' Gee over the favor ain't left.
0:43.4	She knows what she need all I need all she bless a give it my best.
0:48.4	A universal music quickly took it down, but it only drew more attention to the question of how AI is going to affect music.
0:57.0	The voice cloning technology was nothing new. For the past couple months, we've seen it all over Tik-Tok and YouTube. People have been changing
1:04.0	Taylor-Swiss voice to Kanye West. They've tried changing older Paul McCartney's
1:09.3	voice on more recent songs to young Paul McCartney's voice.
1:13.4	They've made Biggie Rap Tupac songs.
1:23.0	All kind of interesting novel
1:27.0	The way up when we rock, and we click with game.
1:30.0	All kind of interesting novelties.
1:32.0	The technology is sort of in its infancy and a lot of the training data that people are using is insufficient.
1:39.0	What you really need is clean these
1:43.6	these models perfectly.
1:44.8	They don't have access to that from people like Coco Band
1:47.2	so a lot of the results are not great. To be clear, a ghost writer who claims their professional songwriter almost
2:01.0	certainly wrote and produced and even sang this song
	...

Please login to see the full transcript.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from Rolling Stone, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Rolling Stone and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.