BEFORE YOU START
Before you start
Prepare an authorized photo and clear speech from one person. The screenshots use a fictional demonstration person and synthetic sample voice.
Screenshots show actual sample projects. Your interface may vary with your plan and updates. Instructions describe desktop browser operation; this guide can also be read on a phone.
Prepare the appearance and voice for one presenter
The goal is a digital double combining appearance and voice. Create the appearance, register the voice, then assign it to that avatar. If you already have either component, continue with the one you have.
Create an original avatar from a photo
Choose a photo in avatar creation, review the name and requested settings, then create the avatar. Inspect it in your avatars, including face, hair and crop boundaries. See the photo-avatar instructions for details.
Register a voice from recording, audio or video
Record the on-screen script or import audio/video in voice registration. Video imports use the audio for training. Use material without music or other speakers, check it and name the voice before registering.
Audition the voice and assign it to your avatar
When registration finishes, audition the voice. Select the avatar you created and assign that voice. Finish the pairing rather than leaving the two components separate, and verify both names.
Add the presenter to a video and test a short line
Add the avatar to a project and check its speech-track voice. Start with a short greeting, listen to the audio, then generate the avatar video. Review appearance, voice and mouth movement together before building the full explanation or training video.
- Action Click Generate ungenerated clips.
- Check After generation, the avatar clip will say Preview generated.
BEFORE YOU FINISH
One final check.
Registering a voice alone does not finish assigning it to the presenter. Verify the pairing and save.