The same speaker, in another language
The source cut and the translated render, playing in step. The voice is cloned from the original, so the delivery survives the language change.
- Source cutTranslated
- Source cutTranslated
One cut in, the same cut in another language
Upload the cut
The finished video, with its audio as recorded. Give it a title so the job is findable in your library.
Choose the output language
More than 175 are supported. One language per run.
Tell it how many speakers
Setting the speaker count lets the model clone each voice separately rather than flattening the room into one.
Render
Five to ten minutes for most cuts, priced per second of output on the tier you pick. The credit figure is on the button before you commit, and failed runs are refunded.
What Translation takes, and what it returns
What Translation needs from you
- Video
- The finished cut, uploaded to your workspace
- Output language
- One of 175+
- Title
- A name for the job in your library
- Typical run time
- 5–10 minutes
Export specifications
- Languages
- 175+ supported output languages
- Output
- Full translated video, or the audio track alone
- Voice
- Cloned from the original speakers
- Delivery
- Rendered cut in your workspace library, re-openable from the History drawer
- Rights
- You own the outputs, to the extent they can be owned, and may use them commercially — see the Terms
Language, speakers, and how the timing bends
The interesting decisions are about fit: a translated line is rarely the same length as the original, and the app gives you both ways of handling that.
-
Output language
175+ languages
One target language per run. The speaker's voice is cloned into it rather than replaced by a stock narrator.
-
Tier
Speed or precision
Speed runs at 154 credits per second of output, precision at 307. Both clone the voice and both cover the same language list.
-
Translate audio only
On or off
Off by default, which returns a full video. On returns the translated audio track by itself.
-
Speaker count
Number of speakers in the cut
Tells the model how many distinct voices to separate and clone.
-
Dynamic duration
On or off
On by default. Lets a translated line stretch or compress to fit the original timing rather than running over the shot.
Engine choice
Two ways to run it. Pick the one that matches your source.
-
Speed
FasterThe faster tier, at 154 credits per second of output.
Volume and drafts
-
Precision
Higher fidelityThe higher-fidelity tier at 307 credits per second, for cuts where the delivery has to hold up.
Client and broadcast work
Built for teams shipping one cut to many markets
-
Course and education teams
Open a library to new markets without re-recording every module.
-
Marketing teams
Localise a hero video without commissioning a voice artist per language.
-
Creators
Reach an audience that does not share your language, still in your own voice.
Often used with
Built for Growth at Every Stage
Every plan includes every model and every feature. Plans only change how many credits you get and how many generations run at once.
- Credits refresh monthly
- Top-up additional credits anytime
- Unused credits don't roll over
Not ready for the commitment?
FAQ
Frequently asked questions
Everything you need to know about Translation on BeHooked.
How many languages are supported?
More than 175. Anything you may have read claiming forty is out of date.
Does it keep the original speaker's voice?
Yes. The audio is produced by cloning the voice in the source cut, so the translated version keeps the speaker's timbre and delivery rather than substituting a stock narrator.
What happens when the translation is longer than the original line?
Dynamic duration, which is on by default, lets the translated line stretch or compress so it still fits the shot. Turn it off if you need the audio to run at its natural pace instead.
Can I get just the audio?
Yes. Turn on translate audio only and the job returns the translated track rather than a rendered video.
What is the difference between the speed and precision tiers?
Cost and fidelity. Speed is 154 credits per second of output; precision is 307 and spends the difference on delivery quality. Both clone the voice and both cover the same language list, so a three-minute cut is roughly 27,720 credits on speed and 55,260 on precision.
What happens if a run fails?
It is refunded. The credit figure is shown on the button before you commit, and a job that does not come back does not stay charged.
