YouTube will today start piloting a new feature that will allow creators on the platform to output short music clips for use in their videos which include AI-generated vocals that imitate specific musicians. The clips will be generated based on text prompts provided by the creator.
Called Dream Track, and backed by Universal Music and Warner Music, the new tool is part of YouTube’s ongoing collaboration with select partners in the music industry to investigate possible uses of generative AI in music.
It is still very much at the experimental stage but looks impressive nonetheless. About 100 US-based creators will have access to the tool, which will allow them to create vocals in the style of artists like Charlie Puth, John Legend, Sia, T-Pain, Demi Lovato, Troye Sivan, Charli XCX, Alec Benjamin and Papoose.
In a blog post, YouTube’s music chief Lyor Cohen and VP Of Emerging Experiences Toni Reid write: “At this initial phase, the experiment is designed to help explore how the technology could be used to create deeper connections between artists and creators, and ultimately, their fans”.
Dream Track is just one of the music AI experiments YouTube is working on following the launch of its AI Music Incubator earlier this year. Other tools being piloted include one that turns humming into an instrumental track and another that can generate a transition between two musical phrases penned by a songwriter.
This all follows the news earlier this week that YouTube is setting up a system via which labels and distributors will be able to get unofficial vocal clones uploaded to its platform taken down.