The exuberant Hollywood actor, Val Kilmer calls himself a “creative soul”, always wanting to explore the world and narrating stories to his fans. Unfortunately, the more expressive star suffered a deploring throat cancer in 2014 and lost his ability to speak after undergoing tracheotomy.
The actor now, sharing his message with an AI custom voice, emerges as an inspiration for both his fans and technology enablers alike.
Recently, Val appeared in record breaking Top Gun: Maverick (2022), which is a sequel to Val’s 1986 Top Gun, continuing his role as Tom the “Iceman”. But as the news surfaced in the media, Top Gun fans appeared confused and wanted to know what would the actor do in this role with no voice of his own?
It’s certainly the most memorable scene of the movie when Tom struggles to talk, dubbed in his AI generated voice. The storyline of Tom is taken from Val’s real life scenario where Tom develops a cancer and communicates using typing, however at some instant in the movie, Val speaks a brief dialogue, in his own voice, crisp and clear, with the same emotions and expressions.
This was possible using AI.
How it all started?
In August 2021, Val Kilmer had announced a partnership with Sonantic– a California based tech startup that specializes in creating AI powered voices, to create his own custom, computer generated voice.
The company makes voice for video games, Hollywood recordings, and speech therapy applications where a voice actor reads out scripts that are provided by Sonantic and are then fed into Sonantic’s pre-developed voice engine.
In the case of Val Kilmer, however, the actor could not record his voice, and Sonantic had to change its approach, thinking out a different, and innovative solution.
Crafting Val’s Voice was not easy!
The team took on the challenge and started with gathering Val’s voice data from his movies, including Top Gun, Batman Forever, The Doors, and Heat. Apart from Val’s movies, he had hours of footage of his 40 year vlogs which were recorded in his personal cameras.
These voice scripts were meticulously cleaned to remove background noise and processed to extract the clear voice of Val. “The scripts were then transcribed and paired with voice in small intervals” says Sonantic.
But, as Sonantic noticed, the data was 10 times less than what normally AI algorithms require to function accurately.
The team then worked on researching and developing a new algorithm that could accurately generate Val’s voice with few data features.
Sonantic remained successful in this endeavor, and Val’s voice was successfully restored just like his original natural voice we get to hear in Top Gun (1986).
Take a glimpse of Sonantic’s AI voice engine, featuring Val Kilmer’s emotional and expressive voice.
Val Kilmer’s daughter Mercedes was reportedly present during the film shootings and told a publishing agency that her father’s role in the Top Gun: Maverick is really “special”.
“They were able to dub him with his own voice, which is amazing. It’s such a technical feat, being able to engineer his voice that way, that it’s an extension of the technical feat of the film.”, – Val’s daughter.
Want to build your first AI model?
Join hundreds of others at Dice Analytics and register for a 2-months short course* on AI and Deep learning.
*Flexible pricing and amazing discounts are available.