Generate audio and sync lips to correct recording (VideoDub)

« To Transcriptions: Overview

VideoDub lets you effortlessly fix mistakes in your recordings without re-recording. Using AI, it automatically adjusts both the audio and lip movements to match your corrected text, making it look and sound like you said it perfectly the first time.

Please note:
This feature is in beta and is being released gradually to Riverside users.

Videodub.gif
Step by step

  1. Log in to your Riverside account.
  2. Open a recording in the editor.
  3. In the transcript, select the word you want to correct.
  4. Click VideoDub.
  5. Type in the replacement text.
  6. Click Generate
    VideoDub generation may take a few moments, but you can continue editing while you wait.

Icon.svg Good to know

  • Currently, VideoDub is only available in English.

  • VideoDub can only be applied to the host’s track. Guest tracks are not supported.
  • You can apply VideoDub to up to two words at a time. Full sentence edits are not supported.

  • Results may vary. AI accuracy depends on factors like pronunciation, lighting, and speaker positioning. For best results, keep your face clearly visible and avoid covering your mouth with your hands, a microphone, or other objects.

Learn more

Did this answer your question?