23 May 2021
A start-up company has a tool that it says can recreate lip sync - matching a speaker's lip movements with pre-recorded audio - in films.
The artificial intelligence, or AI, tool may change how film producers replace recorded speech in a movie with speech recorded in another language. This process is known as dubbing.
Start-up Flawless AI, co-founded by film director Scott Mann, has a tool that it says can recreate lip sync without changing the performance of the actors.
The tool studies how actors move their mouths. It then changes the movements to sync with the replaced, or dubbed, words in different languages. The result could be that Tom Hanks looks like he can speak Japanese or Jack Nicholson looks like he can speak French.
The idea for the tool came to Mann when he saw how dubbing affected his 2015 film "Heist," starring Robert De Niro.
"I hate dubbing as it stands," Mann told Reuters. He went on to describe the problems that dubbing can create. "You have to change so many things to try and catch sync. You're changing words filmmakers and performers have thought about so deeply. They're thrown out to find a different word that fits, but it never really does."
Mann decided to do something about this issue.
After some research, he discovered a paper written by Christian Theobalt of the Max Planck Institute for Informatics in Germany.
The paper lays out a new method that uses artificial intelligence to recreate human visual effects.
The tool worked to translate a De Niro scene in "Heist." Man said the tool was able to take an 'ooh' sound from De Niro 20 minutes earlier and place that in a different moment in the film."
He added, "It measures at the same time and blends it so that the performance is the same, but it's a different mouth movement."
Blend is a term that means to mix something, often with good results.
Flawless AI is now working with producers and studios to use the technology after filming is complete. The first films using this tool are around a year away.
Mann believes this technology will help to remove foreign-language categories at award shows. It could create a more open film industry with a real global stage. "From a filmmaking point of view, you're going to see the rise of a much more diverse range of stars," he said.
Eventually, Mann hopes the technology's use will be widely accepted. "When it's a common thing, we're not going to think about it anymore and we'll just enjoy the movies," he said.
I'm John Russell.
Rollo Ross reported on this story for Reuters. John Russell adapted it for Learning English. Susan Shand was the editor.
Words in This Story
lip-sync – n. to pretend to sing (recorded words)
translate – v. to change words from one language into another language
scene – n. a part of a play, movie, story, etc., in which a particular action or activity occurs
category – n. a group of people or things that are similar in some way
diverse – adj. made up of people or things that are different from each other
range – n. a group or collection of different things or people