Skip to main content
Glass speaking portrait with synchronized audio wave
Open AI Studio → Imagine → Lip sync to create a speaking portrait. This is different from general video generation: the audio supplies the performance that the face follows.

​
Prepare the portrait and voice

1

Choose the portrait

Supply a clear image of the intended face. Use a portrait you own or have permission to animate.
2

Supply the speech

Add an audio reference, or use the integrated Text to Speech option to create narration from your script.
3

Review settings and credits

Choose the supported model/settings and check the estimate before generating.
4

Inspect the result

Play the finished clip. Check timing, pronunciation and facial movement before sharing it.

​
Text to Speech inside Lip sync

The narration option uses the same Speech workspace as the main Speech tab. Choose a voice, model and delivery settings, then use the generated audio for your portrait. See Speech and voices for voice filters, expressive delivery and output settings.
Glass microphone, narration script and sound wave

​
Get a cleaner result

  • Start with a clear face and an uncluttered composition.
  • Use clean audio with one intended speaker.
  • Correct the script and pronunciation before generating the animation.
  • Inspect the full clip; realistic results still need human review.
Yes, supply it as the audio reference and pair it with the portrait.
No. Lip sync starts from a portrait and speech; general scene clips belong in Video.
Speech generation and portrait animation are separate operations. Do not extend the Speech badge to Imagine’s Non-ZDR animation service.
Do not impersonate someone deceptively or animate a person’s likeness or voice without the necessary permission.