Announced September 9, 2026, the feature uses a combination of AI and VFX to adjust on-screen mouth movements so they line up with human-performed dubs — launching globally in English on the first two seasons of Maxton Hall, with more titles to follow.
2026-09-09 · by Moe Ameen
On September 9, 2026, Prime Video announced a new lip-sync capability that adjusts actors' on-screen mouth movements so they line up with dubbed dialogue. The point is to fix the familiar tell of dubbing — a mouth that keeps moving after the translated line has finished, or stops before it — which pulls a viewer out of the story. Amazon says the process is powered by a combination of AI and VFX technologies working behind the scenes.
One detail matters for understanding what this is and isn't: the dubbed audio is still human-performed. This is not AI-cloned voices reading a script — professional voice actors record the dub, and the AI/VFX layer only reshapes the visible mouth to match that performance. Amazon frames the whole thing as running under creative oversight, quoting Raf Soltanovich, VP of technology at Prime Video and Amazon MGM Studios: "With lip-syncing, we're creating a more seamless and immersive way for customers to enjoy global content," done so that "the integrity of their artistic vision is preserved."
The feature debuts on the German romance series Maxton Hall, available globally in English on Seasons 1 and 2 from launch day, and it will be baked into the third and final season when it premieres December 9. Amazon says it is already working to expand the technology to additional titles. As with any newly announced feature, treat the initial title list and rollout timing as a snapshot and confirm against Amazon's own announcement.
The lesson for creators is the gap this leaves, not the feature itself: Prime Video's lip-sync makes global content feel native inside Amazon's own player, but it does nothing for the video you publish everywhere else. If the takeaway is "localized, natural-sounding video is now the expectation," the creator move is to generate portable localized video that actually goes to your platforms. That is where [Kompozy](/) fits — it is a content generation and multi-platform publishing engine, not a streaming feature locked to one catalog. Rather than dubbing over existing footage, its [Persona Shorts](/glossary/persona-shorts) and Persona HeyGen formats generate a talking-head avatar in your own persona that speaks the target language directly through HeyGen's native text-to-speech, so the voice and the mouth are produced together instead of one being retrofitted to the other — the approach behind [AI dubbing](/glossary/ai-dubbing) and [AI lip sync](/glossary/ai-lip-sync) tools, but with the whole asset generated from a [Persona Brief](/glossary/persona-brief) that holds your voice constant across every language.
From there Kompozy does the part Prime Video's feature deliberately isn't for: distribution. One localized talking-head asset feeds [Clipped Shorts](/glossary/clipped-short), captioned vertical cuts, a brand-exact [Carousel](/glossary/hyperframes), Quote Graphics, a Blog Article, and an Email Newsletter, and [Autopilot](/glossary/autopilot) schedules and publishes the batch across eight social platforms plus blog and email. So while a studio uses AI+VFX to sync mouths on a show only its subscribers can watch, an independent creator uses Kompozy to generate a native-language version of their message and put it in front of the audience wherever that audience actually is.
It is a capability Amazon announced on September 9, 2026 that adjusts actors' on-screen mouth movements to line up with dubbed dialogue, so the visible lips match the translated audio instead of drifting out of sync. Amazon says it uses a combination of AI and VFX technologies and applies it under creative oversight. It launched globally in English on the first two seasons of Maxton Hall.
No. The dubbed dialogue is performed by human voice actors; the AI and VFX only reshape the actor's on-screen mouth to match that human performance. So the voice is real and the lip movement is synthetic — the opposite split from a full AI-voice dub. Amazon frames the process as preserving the integrity of the original artistic vision.
At announcement it debuted on the series Maxton Hall — available globally in English on Seasons 1 and 2 — and will be included when the third and final season premieres on December 9. Amazon said it is working to expand the technology to more titles, so the list should grow; confirm current availability against Amazon's own announcement.
Not directly — the feature is built into the Prime Video player for Amazon's catalog, not offered as a creator tool. To get native-language video for YouTube, TikTok, Instagram, and the rest, creators generate it themselves. Kompozy's HeyGen-powered persona video makes a talking-head avatar that speaks the target language directly, holds your voice constant via a Persona Brief, then clips, reformats, and publishes it across the eight social platforms plus blog and email.