The Iceman will not be silenced.
Beloved actor Val Kilmer of Top Gun, Tombstone and Willow fame, among many others, lost his voice -- and his career -- to throat cancer after a tracheotomy in 2014. In 2020, Kilmer turned to British AI company Sonantic to create a model of his voice, basically a custom-built audio-only deepfake for the actor's personal use.
Sonantic creates voice models primarily for video games using actors who read hours worth of scripts. That audio is then fed into the company's voice engine to build a model that can say anything on demand. But Kilmer's case was trickier because there was little audio from his career available in a high-quality format and with minimal background noise.
A team cleaned up the available audio to remove as much noise as possible and attempted to build a model for Kilmer's voice with only about one-tenth of the audio normally required. When the results were subpar, Sonantic developed new algorithms to work with less data and wound up creating over 40 different models of Kilmer's voice. The team chose the best and most expressive one and now you can hear the results below.
Kilmer's voice can be controlled by a desktop application that allows him to type whatever he wants the model to say and then fine tune aspects of the delivery like pitch and pacing.
"The ability to communicate is the core of our existence and the effects from throat cancer have made it difficult for others to understand me," Kilmer said in a written statement. "The chance to tell my story, in a voice that feels authentic and familiar, is an incredibly special gift."
The new voice model wasn't used in the recently released documentary about Kilmer's life, titled Val. That film featured archived footage of Kilmer speaking, as well as the voice of his son, Jack.
While Sonantic's spot-on re-creation of Kilmer's voice is impressive, it again surfaces concerns about AI's constantly advancing ability to create fakery and the potential for misuse and abuse.
There was aearlier this year when it was revealed that another algorithm not related to Sonantic had been used to re-create Anthony Bourdain's voice in a documentary about the late chef and TV star's life. Bourdain's widow was less than thrilled by this revelation, although the film's director Morgan Neville says he got permission from Bourdain's estate.
"We recognize that we have a special responsibility to help prevent the misuse of AI voice technology," Sonantic CEO Zeena Qureshi said in a blog post last month, adding that the company follows the European Union's Ethics Guidelines for Trustworthy Artificial Intelligence. "We do not train algorithms on publicly available data, and we do not create voices where the owner of the voice -- or the owner's estate -- is unaware of its repurpose."
It's not clear where we'll hear Kilmer's new old voice next, but we should get see him in some capacity in Top Gun: Maverick when the long-awaited sequel is finally released this November. The film was originally set for a 2019 debut, but was hit with a year-long delay followed by the inevitable COVID postponement. Now it seems Iceman is ready to fly again, in more ways than one.