YouTube drops a major AI generative software ‘Dream Tracks’ as apart of its ‘Shorts’ featuring Sia, Troy Sivan and CharlieXCX
YouTube has made a groundbreaking foray into artificial intelligence, launching the new generative software “Dream Track” as part of its “Shorts” branch of video content.
This technology, powered by Google DeepMind and the cooperation of nine pop-stars, allows the user to generate a thirty-second song snippet from any text prompt, sung by the ensnared cadences of Troye Sivan, Demi Lovato, Charlie Puth, Charli XCX, John Legend, Alec Benjamin, Sia, T-Pain, or Papoose, over mimetic instrumentation.
With Google DeepMind’s “most advanced music generation technology to date” at its helm, Dream Track was made available to a small selection of artists and content creators just yesterday.
The release’s subsequent blog post, shared by YouTube executives Toni Reid and Lyor Cohen, features two video examples of the technology in action.
The words “A ballad about how opposites attract, upbeat acoustic” are typed above a headshot of Charlie Puth, and a (semblance of a) song comes into immediate being.
The facsimile Puth sings “baby, we’ve got nothing in common – but I know that I’m what you’ve been wanting for so long” over a measured piano and a pair of sorbet cones, displaying how the user might apply sound to sight in their content creation.
Puth has said that he is “extremely excited and inspired” by Dream Track, and asserts with a foreboding choice of words that YouTube “understands the need to work together to develop this technology responsibly, ensuring it will accelerate creativity instead of replacing it.”
The other clip employs the auto-croon of T-Pain, whose extensive artistic use of pitch correction makes him a prime candidate for the software at hand.
Kept in time by the synthesis of a finger-snap and hand-clap, metaphysical T-Pain sings that he “woke up (x3) in a sunshine state” with the misplaced passion of a lustful R&B jam.
Ahead of Dream Track’s grand unveiling, YouTube announced earlier this week that it will introduce “updates that inform viewers when the content they’re seeing is synthetic” within the next few months, and require creators to disclose the same.
“When technological innovation, human imagination, and music meet, extraordinary things can happen […] It’s still early days but we are motivated by our progress in AI and music and are energised by the possibilities that lie ahead.”
Words by Harrison Jones.
