If you have ever seen a Star Wars show on Disney Plus, there''s a good chance you''ll know it or not. Both the Mandalorianand The Book of Boba Fett and an as-yet-unidentified character in Obi-Wan Kenobi have asked Respeecher to keep the name of the character a secret for the time being, and with so many franchise veterans returning, there is certainly no shortage of potential candidates.
Polygon spoke with Respeecher CEO Alex Serdiuk to understand a process that is no doubt about sacrilege: utilizing technology to produce entirely bespoke performances for one (or possibly even two) of the Star Wars saga''s most iconic characters. Serdiuk, from the start, describes the human element behind the platform itself. We also allow [studios] to scale voices, to age, and even resurrect voices for some projects.
Apart from the mental sensation that is conjured up by terms like artificial intelligence and voice cloning that of a sound engineer running lines of dialogue through a computer algorithm that then shuts out audio files Respeechers work on Star Wars is surprisingly performance driven. While Darth Vader himself may be more machine now than man, if the Ukrainian company is supplying his voice (and remember, we said if) the essence of the characters voice is still quite flesh and blood.
There is no AI yet, and I doubt it will exist, that would allow us to use it on a turnkey basis to produce our results.  We need another human voice, as well as all of its inflections, the accent, the speech style, and pace that AI is not good at developing, according to Serdiuk.  It takes all of the performance, all the acting from what we call a source voice, and then we do the conversion.
Aside from the Respeechers pipeline, actors like Hamill can record different takes exactly like they expected on an actual set, which the company''s experts may later adjust at their conclusion based on notes from showrunners like Jon Favreau or Deborah Chow.
If all of these are made up of studio projects and films, we might have to convert them back to the younger voice, and maybe send different versions because we used to train different models with different settings, he says. Can you try to make [a line reading] sound a bit more like they ask? and we might work to make it appear a bit more like they ask.
Is it the ultimate intention to recreate a traditional performance in the case of The Mandalorianor The Book of Boba Fett, as if Hamills lines were somehow transferred directly from the set of 1983s Star Wars: Return of the Jedi? Serdiuk agrees, and the purpose is to make it sound as it was recorded yesterday in the studio.
Even if viewers aren''t sure why, the voice cloning system like Respeechers will sound artificial. Serdiuk even admits that the Hamills de-aged voice is significantly improved in The Book of Boba Fett, as well as in The Mandalorian''s AI model. Despite the fact that many people were concerned about the visual effects used to portray Luke Skywalker in the finale, however, few thought that the Jedi Masters voice was also synthetic until Lucasfilm made the cut in a
Serdiuk mentions that Hamills'' inspiring de-aged vocals from Lucasfilm were much more impressive given the quality of Respeecher''s legacy assets. It''s a matter of acquiring old ADR recordings, something from a video game, and it''s important to get this data trained in your model to produce the output performance that fits into a modern production. In many projects which involve aging or resurrecting, this may be the main obstacle, given that the lack of data and the
The CEO of Respeecher claims that being able to overcome these data-related challenges has been beneficial, now that industry giants like Lucasfilm have embraced the work they do. [We] started with the idea of building a synthetic speech on the level where it would go through sound engineers and Hollywood studios and land in big productions. So when they accept our sound, when they say something good about the sound we were capable of producing, and that''s a very complicated and complex technical challenge, it really encourages us and
Serdiuk believes that a growing acceptance of voice cloning technology might result in that studios no longer need skilled soundalikes to stand-in for deceased actors. For example, would Rogue One: A Star Wars Story in which Guy Henry imitated the voice of Grand Moff Tarkin as the kind of project that would automatically land on Respeechers desk now? Not necessarily, according to Serdiuk, who sees Respeechers voice cloning technology as one of several viable
There are always different perspectives of how things should happen in the industry, and the fans have different opinions. I would not say that [Respeecher] is quite well suited to be in charge or judge, according to Serdiuk. However, since Lucasfilm is a repeat customer, the company has made significant inroads for voice cloning technology.
Serdiuk has already had a vision for the Respeechers future that extends beyond de-aging actors'' voices, although he remains optimistic that what the company has planned will expand filmmakers'' creative horizons, not shrink them. He also talks about democratizing the technology so that smaller film and television companies and video game developers may cut their budgets even further. He also discusses the company''s breakthrough health care improvements, citing one instance where the company is collaborating with a voice actor who has lost
Serdiuk has lost sight of what it takes for Respeecher to play a certain space opera set a long time ago in a galaxy far, far away. It''s something special. I mean, this story is part of the story. From a technical perspective, they have been disrupting the entertainment industry, and the way they do their films is outstanding. So, it''s a huge pleasure to be able to collaborate with those individuals and learn from them.