Introducing Dubbing v2

Dubbing v2 is our most natural-sounding dubbing model yet — preserving the emotion, timing and identity of the original speaker while translating across more than thirty languages.

The new model dramatically improves lip-timing alignment and handles overlapping speakers, music beds and background ambience without bleeding artifacts into the dubbed track.

Studios can now turn around localized cuts in hours instead of weeks, with full control over per-segment voices and a review workflow built for teams.

Dubbing v2 is rolling out to all IntuneLabs accounts this week, with API access available for enterprise customers.