AI-Generated Music: AI music generators, like Suno or Udio, have noticeable tells, much like AI image generators.
Details: These tells might become desirable in the future, similar to how people appreciate the imperfections of older technologies. One example is how Suno and Udio are trained on low-quality MP3s, resulting in a grainy, hissy sound reminiscent of AM radio or early digital music formats.
Horns Analogy: A key aspect of real music lies in its imperfections. For instance, the beauty of a horn section comes from the slight variations in timing and pitch among the different horns.
Details: This aspect is currently challenging for AI music generators to replicate effectively, as they tend to prioritize perfect synchronization and pitch.
00:00
Transcript
Episode notes
For the second episode in our three-part miniseries about the future of music, Charlie Harding, a music journalist and co-host of the Switched on Pop podcast, joins the show to tell the story of Auto-Tune. He walks us through how a simple plugin became such a recognizable sound in music, why both artists and fans gravitated to the Auto-Tune sound, and why Auto-Tune has continued to grow even through backlash in the music business. Then we look ahead to AI, and try to figure out what — if any — lessons we might be able to learn about the sound and culture of the AI era to come.