Prime Video Is Now Using AI to Change Actors’ Mouth Movements in TV Shows

Prime Video has introduced new AI-assisted technology that can change an actor’s mouth movements so dubbed dialogue appears to match what they are saying on screen.

The feature has launched on the English-language versions of Seasons 1 and 2 of German drama Maxton Hall, with the technology also confirmed for the third and final season when it arrives on 9 December 2026. Prime Video says it plans to expand the system to more titles in future. 

Traditional dubbing can create an obvious mismatch between the words viewers hear and the movement of an actor’s mouth, particularly when dialogue has been translated from another language. Prime Video’s new system uses artificial intelligence alongside visual effects to adjust those mouth movements so they better match the human-recorded English dialogue.

Importantly, this does not mean AI is replacing the actors providing the dubbed voices. Prime Video says the audio is still performed by human voice actors, while the AI is being used to alter the picture and make the translated performance look more natural. 

Prime Video says the aim is to make international programmes easier to watch without viewers being distracted by dialogue that does not visually match the actors on screen. The company also says the process is carried out with creative oversight so the original performance remains central to the finished result. 

Maxton Hall is a particularly high-profile place to introduce the technology. Amazon describes the series as its most successful scripted international Original, with Harriet Herbig-Matten and Damian Hardung returning for its final season later this year. 

For viewers, the change could eventually become one of the most noticeable uses of AI in everyday streaming. If Prime Video expands it across more international films and series, dubbed versions could begin looking much closer to performances originally recorded in English.

The technology also shows how streaming services are increasingly using AI in ways that audiences may not immediately realise, with changes being made directly to finished television footage rather than only behind the scenes during production.