A video posted to Twitter on Thursday suggests that he just dropped a new sound from rapper Jay-Z. Except that the 40 seconds of sound does not come from the New York rapper but from an artificial intelligence, in an almost perfect imitation.
The project comes from Alltamusic, two French DJs. On Instagram they explain that they used “an artificial intelligence filter” to reproduce Jay Z’s voice, over a text written and performed by one of the duo’s artists.
precedents
This is not the first time that the American rapper’s voice has been artificially reproduced.
Already in 2020, the creator of the YouTube channel Vocal Synthesis had managed, using the open source artificial intelligence software Tacotron 2, to get Jay Z to read a text by Shakespeare. The result was quite convincing.
In 2021, the YouTube channel 30 Hertz also published a sound of just 1 minute, where Jay-Z and Kanye West rapped together. The creator specified in the introduction of the content that “all the voices in this video are generated entirely by an AI.”
Jay Z is not the only artist artificially imitated. Travis Scott or Kendrick Lamar had the right to reproduce his voice. Content creator Gengarcade even had fun rapping Drake about his hatred for beans, using ChatGPT to write the verse and software UberDuck for the vocals.
legal issues
In 2021, Jay Z’s label attempted to have videos made by Vocal Synthesis removed from YouTube for copyright infringement. The creator of the YouTube channel had then received a surprising notification: “This content is illegally using an AI to imitate the voice of our client.” The video had been removed for a while, before being made available again today.
The copyright debate surrounding works created by artificial intelligence is regularly in the news.
“The future of music”
In February 2023, French DJ David Guetta sparked a mini-controversy. he had broadcast during one of his concerts a verse by American rapper Eminemcreated by the DJ himself through artificial intelligence software.
Many had wondered about the dangers of appropriating a vocal identity. The DJ had responded in a BBC interview that he did it as “a joke that worked well”. However, he explained that he was “sure that the future of music lies in artificial intelligence. But as a tool.
Source: BFM TV
