Audio-only translation and Enhanced Cloning
How to translate only the speech in a video, get a dubbed video without lip sync, and when to use Enhanced Cloning.
Written By Umakhan Magomedov
Last updated About 6 hours ago
Translate Video always creates a new dubbed video. If you only need the translated speech, or a translation without lip sync, there are simpler and cheaper options. This article also explains Enhanced Cloning.
Translate only the audio of a video
Use Translate Audio. It accepts video files, takes the sound from the video and gives you the recognized text, the translation and a voiceover of the translation.
Open Translate Audio from the Tools tab.
Tap to choose a file and select the video itself. You don't need to extract the audio first.
Wait while VocaLingo takes the sound from the video. On older phones this can take a few seconds, so keep the screen open while the loader is shown.
Choose the target language. Recognition and translation start automatically.
On the Translation tab, tap play to generate the voiceover. You can listen to it and share it.
💡 Already opened the video in Translate Video? Tap Translate audio under What next? to send the same video to Translate Audio.
Translate Audio is charged per step and costs far less than dubbing. A one-minute recording with voiceover usually costs from about 1 to about 20 tokens, depending on the voice provider. Details: Token pricing for each tool.
Dubbed video without lip sync
Translate Video no longer has a separate Audio only switch. To get a dubbed video without lip sync, open Settings in Translate Video and choose ElevenLabs. It clones each speaker, keeps the original music and sound effects and doesn't change the picture. Lip sync for ElevenLabs is an optional extra.
Enhanced Cloning
Enhanced Cloning is a HeyGen option. It uses a more accurate voice model, so the dubbed voice is closer to the original in tone and character. It costs twice as much as standard HeyGen.
Use it when:
Voice authenticity matters: interviews, personal content, documentaries.
The speaker has a distinctive voice you want to keep in the translation.
The speaker is on camera and viewers know the original voice.
To turn it on, open Settings in Translate Video, choose HeyGen and enable Enhanced cloning. The token estimate under the Translate button updates right away.
Which option to choose
Frequently asked questions
Where is the Audio only switch?
Where is the Audio only switch?
It was removed from Translate Video. For a dubbed video without lip sync, choose ElevenLabs. For the translated speech only, use Translate Audio.
I have tokens, but the app asks me to buy more for a video. Why?
I have tokens, but the app asks me to buy more for a video. Why?
Translate Video charges for every second of video, about 3.67 tokens per second, so 80 tokens cover about 20 seconds. The estimate is shown before you start. If you only need the speech translated, Translate Audio is much cheaper.
Does Enhanced Cloning work with ElevenLabs?
Does Enhanced Cloning work with ElevenLabs?
No. It's a HeyGen option. ElevenLabs clones each speaker by default.
Can I save the translated audio as a file?
Can I save the translated audio as a file?
Yes. In Translate Audio, generate the voiceover on the Translation tab and use the share button in the player to send or save the file.
Related articles
Was this helpful?
Still need help? Share an idea