YouTube unveiled its TikTok challenging “Dream Track” early Thursday, a tool that uses artificial intelligence to generate soundtracks and lyrics in the styles of nine recording artists, including Charlie Puth, Demi Lovato, John Legend and T-Pain.
Dream Track is powered by Lyria, an AI-music generator developed by YouTube parent Google’s DeepMind labs. For now, it is being offered to what the company called “a small group of select US creators” to give them a chance to create unique soundtracks up to 30 seconds longs for the short videos they post on the site.
Two samples of the potential were offered, one featuring R&B artist T-Pain created with the prompt, “A sunny morning in Florida, R&B.”
“I woke up in the sunshine state!” the result sings out over a video of a woman walking on the beach at sunrise.
The second video prompt, “A ballad about how opposites attract, upbeat acoustic,” quickly returns Charlie Puth’s voice singing, “Baby we’ve got nothing in common…”
YouTube explained in an introductory video that creators can use Dream Track to create unique soundtracks by simply typing in an idea, or even humming a melody and prompting the AI to use it to create a saxophone track, for example.
“The potential of AI is incredibly exciting. But with any new technology, we have to approach it responsibly. And that’s not lost on us,” said Lyor Cohen, global head of music for YouTube and Google, in the video. “It starts by recognizing that artificial intelligence is meant to amplify human activity, not replace it.”
The Dream Track experiment is designed to help explore how the technology could be used to create deeper connections between artists and creators, and ultimately, their fans,” Cohen and Toni Reid, VP of emerging experience and community projects, said in a blog post accompanying the video.
The reveal comes amid rising concern about the use of AI to mimic artists’ voices. Earlier this year, an AI-generated song that sounded like Drake and The Weeknd caused a stir when it was submitted for the Grammy Awards and at first considered eligible for an award before the Recording Academy walked that back.
But it’s also being used in creative ways, including the production of what Paul McCartney called the “last” Beatles song, “Now and Then,” created using an old cassette recording of John Lennon enhanced with AI.
YouTube said last week it will require disclosure of AI-generated material by its creators.
In addition to Puth, Lovato, Legend, and T-Pain, the lineup of collaborating artists from multiple genres includes Alec Benjamin, Charli XCX, Papoose, Sia, and Troye Sivan. “By simply typing an idea into the creation prompt and selecting a participating artist that appears in the carousel, an original Shorts soundtrack featuring the AI-generated voice of that artist will be produced for the creator to use in their Short,” Cohen and Reid said in the blog post.
“YouTube has been a great partner in outlining their approach to AI and understands the need to work together to develop this technology responsibly, ensuring it will accelerate creativity instead of replacing it,” Puth said in a statement accompanying the announcement.
“When I was first approached by YouTube I was cautious and still am, AI is going to transform the world and the music industry in ways we do not yet fully understand,” Charli XCX said. “This experiment will offer a small insight into the creative opportunities that could be possible and I’m interested to see what comes out of it.”
Lovato said that artists “need to be a part of shaping what the future looks like.”
YouTube called the reveal of Dream Track and “sneak peak” and said it continues to explore the possibilities and challenges of AI, pointing to the AI Music principles it unveiled in August. “By testing and learning, we can better understand how to enhance artist and industry opportunity and deepen fan engagement, benefiting the broader music community,” the blog post said.