YouTube Shorts will integrate Dream Track, an experimental feature that allows you to create soundtracks with the voices of 9 international artists generated by AI of up to 30 seconds for their short films.
Even though lately Microsoft and OpenAI hog the spotlight in the spotlight with the ever-deeper integration of Copilot AI in Windows 11, Google nonetheless remains a reference in the scientific community in terms of artificial intelligence. A month ago, we told you about a project from the search giant which would make it possible to recording yourself with the voice of an existing singer.
This voice replacement tool and music generation powered by AI is called Dream Track and it is accessible in YouTube Shorts. This experiment is the result of the work of Google DeepMind, which designed an advanced music generation model called Lyria. The latter does not just replace your voice with that of famous artists such as Alec Benjamin, Charlie Puth, Charli XCX, Demi Lovato, John Legend, Papoose, Sia, T-Pain or even Troye Sivan.
Dream Track from YouTube will let you sing along with the voice of your favorite artist
Lyria, the model used by Dream Track on YouTube Shorts, also allows you to replace a vocal recording with the instrument of your choice, the trumpet, for example, as can be seen in the demonstration video published on YouTube. Lyria will allow everyone to ” create a guitar riff just by humming it or take a pop song you’re working on and give it a reggaeton feel”.
This feature is unfortunately still in testing phase on a “small group of American creators”, the latter being currently only able to create “soundtracks of a maximum duration of 30 seconds for their short films”. That said, there is no doubt that this feature will be a huge success with all Internet users as it promises to be simple to use. In fact, you just need to enter a text describing the expected sound result“a ride on how opposites attract,” for example, and to choose the artist who will lend you their voice in the carousel.