How A Deepfake Entirely Replaced Mark Hamill’s Luke Skywalker

Because One Day We’ll All Be Sky Walkers on The Cloud

There’s something to be said about the rise of the Deepfake. Technology that emulates a person’s real-life characteristics so well that it’s become harder to tell what’s real versus fiction. The scrutiny of this synthetic media technology has been a topic of interest for the past few years. First, with pornography, and then with cybersecurity, as social media applications such as Instagram can be hacked using digital video deepfake recreation of a person.

To see just advanced it has gotten, most recently, the Museum of The Moving Image featured Deepfake clips from an Emmy-award-winning documentary called In Event of Moon Disaster. In it, guests were greeted by a clip of Richard Nixon talking about the Apollo moon disaster. 

 

Except it wasn’t Nixon. Just like how Luke Skywalker in the recent Star Wars TV Universe, are recreations created by digital technology. Last year’s Mandalorian finale – talk of the internet buzz as it was – featured a fully digitally recreated young Luke Skywalker battling a series of battle droids. In fact, most Marvel movies now use a technique called previsualization which both digitally storyboards, acts out, and emulates an actor’s lines and movements way before a shot is even filmed. 

In a recent report by Esquire, details about the CGI Luke Skywalker had been revealed. Alarmingly, it was revealed that not only did they recreate Mark Hamill using digital copies overlaid on an actual actor but apparently, they had even completely reconstructed Mark Hamill’s voice using AI. Completely omitting Mark from having anything to do with his own cameo.

“Something people didn’t realize is that his voice isn’t real. His voice, the young Luke Skywalker voice, is completely synthesized using an application called Respeecher,” said Director Jon Favreau in Disney Gallery: Star Wars: The Mandalorian.

Respeecher is a futuristic sound editing software used to recreate virtually any sort of speech. A neural network that’s fed information to recreate a person’s voice.  

Sound editor, Matthew Wood, shared in regards to Mark Hamill’s cameo, “I had archival material from Mark in that era. We had clean recorded ADR from the original films, a book on tape he’d done from those eras, and then also Star Wars radio plays he had done back in that time. I was able to get clean recordings of that, feed it into the system, and they were able to slice it up and feed their neural network to learn this data.”

BoJack Horseman once did a season-long parody where the eponymous horse wasn’t even present to shoot the movie Secretariat, in which he was the lead. Yet it was still never-the-less nominated for an Oscar in the show. We’ve officially entered that absurd Hollywood reality. What it means for actors, only time will tell. 

 

Christian Angeles
Christian Angeles
Christian Angeles is a screenwriter who likes sharing stories and getting to meet people. He also listens to words on the page via audible and tries to write in ways that make people feel things. All on a laptop. Sometimes from an app on his phone.

Latest articles

Related articles

This site uses Akismet to reduce spam. Learn how your comment data is processed.