Generative AI can now turn snippets of a person’s voice into an entire song. Of course, artists and music labels object to this practice, believing it to be copyright infringement.
However, this technology makes it possible to bring the voice of a deceased artist back to life. And the Beatles take advantage of it.
On BBC’s Best of Today, Paul McCartney shared his thoughts on AI in the music industry and revealed that The Beatles will soon be releasing a new track with the help of AI. “When we started recording what will be the last Beatles record, we worked on a John demo,” said Paul McCartney. “We were able to take John’s vocals and make it pure through AI, which allowed us to mix the album like we normally would.
Rumors About Lennon’s Unfinished Song, “Now and Then”
Without revealing the name of the demo, he said the recording had just been completed and would be released this year. The demo McCartney is rumored to be referring to is Lennon’s unfinished song, “Now and Then”, which had previously been mooted as a possible reunion song.
This demo-to-song technology is not new to McCartney or the Beatles, as it has been used for one of their projects in the past. “We were able to use this type of technology when Peter Jackson made the movie ‘Get Back’, in which we were preparing the Let it Be album, and he was able to extract John’s voice from a tape of poor quality,” McCartney said.
Using AI to produce the band’s latest track, McCartney acknowledged, is “a bit scary but exciting because it’s the future.”
AI music generation continues to grow. Last week, Meta unveiled MusicGen, an open-source music generation tool that can take text or melody and output it to audio. Other models that perform well in this area include OpenAI’s JukeBox, Google’s Music LM, and Riffusion.
Learn more about artificial intelligence and music generation
Source: “ZDNet.com”