Dubbing

Dubbing translates a video or recording into other languages. Each speaker gets a matched voice, the timing follows the original, and the music and sound stay.

Your dubs

Dubbing opens on your dubs, newest first, with each one's target language, version, credits, length and when it was made. Search them by file name or language, and choose one to open it.

Dub a file

1. Open Dubbing in the sidebar.
2. Drop a video or recording anywhere on the list, choose Choose file in the bar at the top, or choose New dub (top right).
3. In the New dub window, check the file. Under Dub into, check the language. Spanish is picked to start.
   • Choose + Add a language to add more.
   • Choose × on a language to remove it.
4. Choose Dub into language (or Dub into N languages). The estimate shows beside it.

The file uploads, then the dub opens and starts. You see each stage, for example Transcribing and finding the speakers…, Translating… and Recording the new voices…. A minute of video takes a few minutes. You can leave the page; it keeps going while it's open somewhere, and picks up when you come back.

How it works:

• The original language is detected automatically.
• Each language is its own dub, with its own files.
• Each speaker is given the closest Cinara voice by gender and age. Nobody's voice is cloned.
• Lines are translated to fit their timing, and shortened where the new language runs long.

The result

The dub fits the window: its details and a player at the top, the lines below (they scroll on their own), and a panel on the right with Voices and Downloads tabs. All dubs (top right) goes back to the list.

At the top you see the languages (for example "EN → Spanish"), the length and the credits.

• Above the player, switch between Dubbed video (language) (for video files), Dubbed audio (language), New voices only (just the new speech, without the background) and Original.
• Voices: each speaker, the voice they were given, and their gender and age, for example "Speaker 1 · Mentor · Nathan · male, adult".
• Downloads: each file, plus SRT subtitles and VTT subtitles in the new language.
• Lines: a table with Time, Speaker, Original, the translation, and Fit. Fit shows how long the new line is compared with the room it has, for example "2.1 of 2.4 s". It also notes when a line was sped up (×1.1) or shortened.

Edit a dub

1. Choose Edit dub.
2. Change any translation in the Lines table.
3. Or, in the Voices tab, choose a different voice for a speaker. The play button beside it plays a preview; choose the voice to search and hear every voice.
4. Lines that will change are marked will be said again. The bar at the top shows how many lines and about how many credits.
5. Choose Apply changes. Choose Cancel to discard your edits.

Only the changed lines are said again, then everything is mixed into a new version. The earlier version stays as it is. The new version shows version 2 and a link to the earlier version.

• Your own wording is kept. If it's too long, it's sped up to fit, never shortened.
• Changing a speaker's voice says all of that speaker's lines again.

Limits

• Files up to 10 minutes and 2 GB.
• MP4 or MOV video, or an audio file.
• One or more speakers.
• Up to 5 languages at once.
• 37 target languages (the same list as Speech).
• Edited lines up to 1,000 characters, and a line can't be empty.

Credits

• 2,000 credits per minute of the original file, for each language, rounded up.
• The minimum is 100 credits per language.
• For several languages, the page shows the total and the price for each.
• Edits: 1 credit per character of the lines said again, plus 100 credits for the new mix.
• If a dub fails, its credits are refunded.

Tips

• Use clear recordings with little overlap between speakers.
• Check the Fit column. Lines marked in red run long; shorten the translation or pick a faster-speaking voice.
• Download New voices only if you want to mix the audio yourself.
• Try one language first, check it, then add more.

Related

• Transcribe: just the transcript.
• Voice Library: find voices for your speakers.
• Clean Audio: split speech and background yourself.
• Video