Hand it a clip of someone talking and pick a language. The speech is translated, spoken back in a voice that fits, and the mouth is re-synced to match it.
Translation and lip sync in one pass
Dubbing usually means three jobs: transcribe, translate, then re-record and hope it fits the picture. Here it is one run. The words are carried across and the mouth is rebuilt around the new audio.
Thirteen-plus languages, chosen from a list so a typo cannot cost you a run
The speaker keeps their delivery and timing, so it still reads as their performance
Optional burnt-in captions in the target language on some models
One shoot, every market
The clearest win is reach: a single piece of footage becomes the same ad, explainer or course module in every language you sell in, without booking talent again.
Ads and product videos for more than one market
Course and onboarding content localised without a re-record
Feed the result to auto-captions for subtitles that match the new audio
Questions
Does dubbing change the speaker's voice?
The models differ, and the picker says which is which. Some carry the original voice across to the new language and others use a matched voice. Either way the delivery and timing follow the original take.
Is the mouth re-animated too?
Yes, and that is what separates this from laying a translated track over the picture. The speaker's mouth is rebuilt around the new audio, so it does not read as dubbed.
Which languages can I dub into?
Spanish, French, German, Italian, Portuguese, Dutch, Polish, Turkish, Hindi, Japanese, Korean, Chinese and English, depending on the model. You pick from a list rather than typing a code.
How is dubbing priced?
By the length of the clip. One model bills per started minute, so a very short clip costs the same as a one-minute one. The token price is quoted before you run either way.
Can I re-voice a clip without translating it?
Yes, with the lip sync task. Give it any audio track, in any language, and it syncs the mouth to whatever you hand over.