Skip to main content

AI Voiceover

The Voiceover node generates speech from text using ElevenLabs v3 text-to-speech. Choose from 21 voice presets to narrate your video.

Inputs

Outputs

Available engines

Voice presets

21 voices are available: Each voice has a preview you can listen to before selecting.

How it works

  1. Receives a script (text) and a video from input handles
  2. Select a voice from the preset library
  3. Preview the voice on your script
  4. Generate the voiceover
  5. The audio is combined with the video and output

Configuration options

For API runs, pass under model_overrides[step]: Unrecognized voice IDs are silently dropped — the rest of the run completes but the voiceover track is missing from the final mix, so double-check the spelling.

Credit cost

~3 credits per 1,000 characters of script text. Cost scales with the length of your voiceover script.

Tips

  • This node is interactive — you select the voice and preview before generating
  • Listen to voice previews to find the best match for your content
  • The voiceover is timed to the video duration
  • Place this before Approve & Export to add narration to your final video

Example use cases

  • Narrating explainer videos
  • Adding voiceover to motivational content
  • Creating talking-head style content without recording audio