The voice of the character from Top Gun: Maverick was made using artificial intelligence, and you did not notice it

Artificial intelligence is increasingly playing a leading role in the audiovisual world. And not just as an active part of the story – as if it were a character -; as well as a key part in its production. The latest example of this is the successful Top Shooter: Maverickwhere artificial intelligence being used to recreate Val Kilmer’s voiceAmerican actor who played Tom “Iceman” Kazansky.

This actor, remember he lost the ability to speak due to throat cancer he was battling in 2014. The hardships that the film covers, as the character played by Kilmer also suffers from cancer that leaves him unable to speak.

However, at some point in Top Shooter: Maverick, Val Kilmer says a few words. This is where artificial intelligence comes into play.

How does the artificial intelligence that voices Val Kilmer in Top Gun: Maverick work?

The key to this process is found in Val Kilmer’s old tapes. The company Sonantic, with which the actor was linked in 2021, took these fragments to create a certain model of his voice and train algorithms that make possible what is seen in Top Shooter: Maverick.

“We have created over 40 different voice models and have chosen the best, highest quality and most expressive,” Sonantic’s CTO and co-founder said in a statement to luck. However, getting to this place was not easy. The database with which they had to train the machine learning system was about ten times smaller than usual.therefore, it was necessary to develop new algorithms capable of achieving voice matching despite this limitation.

“As humans, the ability to communicate is at the core of our existence and the side effects of throat cancer make it very difficult for others to understand me. Being able to tell your story in a voice that feels authentic and familiar is a particularly amazing gift,” said the actor from Top Shooter: Maverick at the time he agreed to work hand in hand with Sonantic.

This is not the first time that artificial intelligence has been used to recreate the voice of another person. popular podcast hkingfrom Spotify too used this technique to recreate the voice of Francisco Franco and tell some fragments of certain episodes.