Video Translator with AI – Dub into 12 Languages
Chat shortcut: /dubbing
Video translator in one step: upload a file, choose a target language, download the dubbed version. The original voice is recreated in the new language, music and sounds are separated and mixed back in; the lips in the picture are not changed.
With Account
Cost: 111,200 tokens per minute or part thereof · Input and result are automatically deleted 24 hours after upload.
This tool can be used with a free account. Sign up for free
One recording, a new audience
An explainer video, a course or a product video is finished, but only in one language. Re-recording a second version costs studio time and a narrator. AI dubbing translates what is spoken and re-speaks it in the target language, with a voice modeled on the one in your recording.
This is how it works: the tool recognizes the spoken language and transcribes it, translates the text, separates voice and background sound and generates the new speech track in the timing of the original. Music and sounds are mixed back in; small deviations from the original mix are possible; the picture stays as it was. You get the recording with the new audio track for download.
If you only need the written text, create a transcript in the original language with "Caption Generator" and translate it afterwards, for example with the translator in the chat; mouth movements matching the new language are not available here. For explainer videos, interviews, talks, courses and podcast episodes where the voice is in the foreground, dubbing is the short route to a second language version.
How it works
- 1
Upload a file
Drop or select video or audio (MP4, MOV, MP3, M4A, WAV and more) from 2 seconds to 10 minutes and up to 96 MB.
- 2
Choose languages
Set the target language from twelve languages; have the source language detected automatically or set it yourself.
- 3
Dub and download
Check the price, click "Let's go", listen to the version and save it, or pick it up later under "My results".
Good to know
- Input: video or audio (MP4, MOV, MP3, M4A, WAV and more), 2 seconds to 10 minutes and up to 96 MB per file. Output: video with a new speech track, with audio input an MP3.
- Twelve target languages, source language automatic or selectable from nine languages; the original voice is recreated, background sound and music are separated and mixed back in.
- No lip sync: the picture stays unchanged, only the audio is replaced.
- Billed per started minute in tokens, amount visible before you start and debited at the start; account only; on technical failure it is credited back, a finished but unsuitable result counts as a run.
- Processing via our server and an external service for speech generation, whose terms govern its storage; on our side, input and video result are deleted 24 hours after upload (hourly cleanup), audio results stay in the media library.
- Chat shortcut: /dubbing.
Frequently asked questions
How good is the result?
With clearly spoken explainer videos, interviews and talks, the new version usually sounds like the original voice in the other language and follows the timing of the speech. Music under the voice, overlaps, strong accents and many speakers lower the quality; technical terms and proper names can be mistranslated. Listen to the whole version before publishing.
Are the lip movements adjusted?
No. Only the audio track is replaced; the picture stays unchanged. With a voice-over this goes unnoticed; with close-ups of a speaker, mouth and sound visibly do not match.
How do I prepare my video?
A clean voice recording without music, reverb or background noise, one speaker at a time, short sentences: translations often come out longer than the original.
Which languages are supported?
Target languages: English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Turkish, Japanese, Chinese and Arabic. The source language is detected automatically or you set it, for example German, to translate a video into English.
How long does it take and what does it cost?
The time varies with length and load; you can pick up the version later under "My results". Billing is per started minute in tokens; the amount is shown before you start and debited at the start; account only, 2 seconds to 10 minutes and 96 MB per file. If a job fails technically, the tokens are credited back.
What happens to my file?
It is handed to our external service for speech generation for dubbing; what it stores is governed by its terms. On our side, input and video result are deleted 24 hours after upload; an audio result stays in your media library.
Can I have any voice recreated?
Only your own or one whose speaker has agreed. For other people's videos you need the permission of the rights holders, also for the translated version.