All products

Dub a video with AI in ElevenLabs

Create a project from your own media, add a target language and retrieve its completed audio. Review the source transcript first when your workspace permits edits.

To dub a video with AI in ElevenLabs, give the dubbing skill a recording you have permission to use and the language your audience needs. A training team might start with a single product walkthrough. The agent can prepare the source once, then add separate target languages to that project.

Product names deserve an early check. If a word was misheard in the source transcript, the mistake can carry into more than one translation. Enterprise workspaces can edit segments and regenerate affected language outputs. Project creation, adding a language and retrieving completed audio are distinct stages, so each version can be checked separately.

What it does

  • Create a project from an authorised audio or video source.
  • Inspect the source transcript before adding languages.
  • Request separate language versions in the same project.
  • Check whether a particular language has finished processing.
  • Retrieve the finished audio for review alongside the original.

Install and use

  1. Install the skill and official ElevenLabs CLI or SDK. Configure an authorised API key and sufficient credits. Transcript edits and regeneration require an Enterprise workspace.
  2. Ask “Prepare this walkthrough and show the source transcript before creating our second-language audio.”
  3. Review product names and timing, then request and check the completed language output.

FAQ

Source

Made by ElevenLabs. Original skill: dubbing, published on GitHub under the MIT licence.

View on GitHub