The world’s largest record label offers singers artificial voice cloning

The world’s largest record label, Universal Music Group, announced plans to offer its artists AI models of their own voices.

The group’s deal with digital startup SoundLabs will see the creation of software called MicDrop that will recreate an individual voice using machine learning techniques. The artist will retain full control over the model – it will not be used without permission, will not be publicly available, and the owner of the voice will profit from its use.

MidDrop is the latest generation of a technology that has been in development for years and is already having an impact on the music industry. Practical applications include being able to sing in foreign languages ​​without having learned them; being able to make music when health problems prevent the vocal cords from working; and even being able to make music after death.

READ MORE: Listen to 25 AI covers of James Hetfield: Toto, Wham!, Seal and more

In addition, the voice model can be used like the voices of a synthesizer, offering additional creative possibilities. While its use as a stage instrument is not yet known, it could theoretically be used by a singer like Jon Bon Jovi – who fears he may never be able to tour again due to health reasons – to stay on the road.

“MicDrop is the first in a series of interoperable AI tools and services developed by SoundLabs for sound design and music generation,” UMG said in a statement. “It gives artists new ‘music superpowers’ and completely reinvents the way music is made, allowing them to expand their possibilities.”

“The goal of SoundLabs is to give artists powerful new composition tools while helping them properly manage their intellectual property. SoundLabs … is focused on helping artists maintain creative control over their data and models.”

AI developer: The future of music belongs to humans

“We believe the future of music creation is distinctly human,” said BT, a musician and head of SoundLabs, which uses the slogan “Transform your voice into any voice you want, instantly.” He continued: “Artificial intelligence, when used ethically and trained consensually, has the Promethean ability to unlock unimaginable new creative insights, reduce friction in the creative process, and democratize creativity… We are developing tools not to replace human artists, but to amplify human creativity.”

Chris Horton, SVP of UMG – which recently announced its principles for music creation with AI – added: “UMG is committed to putting artists at the center of our AI strategy so that technology is used in service of art, not the other way around.

“We are thrilled to be working with SoundLabs and BT, who have a deep and personal understanding of the technical and ethical issues surrounding AI. SoundLabs will enable UMG artists to push creative boundaries by using voice-to-voice AI to sing in languages ​​they don’t speak, perform duets with their younger selves, restore flawed vocal recordings, and more.”