How the voice of Darth Vader will be preserved by artificial intelligence

Darth Vader.Star Wars Series

James Earl Jones approved Disney’s use of Respeecher, an AI voice-modeling program, to recreate his vocal performance as Darth Vader during the development of the Obi-Wan Kenobi TV series, according to a Vanity Fair article published on Friday.

The renowned Star Wars villain has been voiced by Jones, 91, for 45 years, beginning with 1977’s Star Wars: Episode IV—A New Hope and ending with a single line of dialogue in 2019’s The Rise of Skywalker. In an interview with Vanity Fair, Matthew Wood, a supervisory sound editor at Lucasfilm, stated, “He had hinted he was looking towards winding down this specific character.” “So, how do we proceed?”

Respeecher, a voice cloning device from a firm in Ukraine, provided the solution. It models and replicates human voices using deep learning in an almost imperceptibly similar way to the original. Lucasfilm previously utilized Respeecher to duplicate Mark Hamill’s voice for The Mandalorian, and the studio believed the same technology would be excellent for a significant Darth Vader cameo requiring dozens of lines of dialogue. Using Jones’ vintage recordings as a starting point, speech-to-speech business Respeecher developed a voice model that could be “played” verbally by another performer.

How the voice of Darth Vader will be preserved by artificial intelligence 2

Russia’s invasion of Ukraine occurred while the play was under development, providing numerous terrifying difficulties for the Respeecher crew, as Vanity Fair examines in its in-depth reported article. Respeecher is allegedly working on additional top-secret projects despite the conflict, potentially for Disney, so more Darth Vader is to come.

Respeecher is aware that, putting Vader aside, this sort of technology might create serious issues with copyright, social engineering, and security if anybody can duplicate another person’s voice using recorded samples. According to the business’s ethical statement, the company “does not enable any fraudulent uses of our technology” and “does not employ voices without consent where this can affect the subject’s privacy or their capacity to make a livelihood.”

However, Respeecher’s laboratories won’t always be the only ones using voice cloning technology. As we noted earlier last month, initiatives like Koe Recast are helping the open-source community catch up. It’s probably just a matter of time before this powerful genie bursts free permanently. Then, we will all resemble Darth Vader.

Exit mobile version