Help Center

Digital Human

The Digital Human feature generates presenter videos with your cloned face and voice: upload a short video to clone your avatar, record a voice sample to clone your voice, then type a script or upload audio to produce the video. Find it under "Digital Human" in the sidebar.

Two Drive Modes

Pick the drive mode before creating a video. Output quality is the same — the difference is where the voice comes from.

Text drive
Type your script; AI reads it with your cloned voice and drives the avatar's mouth and expressions. Great for batch voiceover videos — just edit the text and regenerate.
Audio drive
Upload a real recording (MP3/M4A/WAV, up to 20MB, 5s–3min) and the avatar speaks along with it. Ideal when you want your own natural voice.

Clone Avatar & Voice

Create your avatar and voice once — each takes a few minutes and is reusable forever.

  1. 1Click "Create" next to "Select Avatar" to open the video-cloning page.
  2. 2Enter a name and upload a clear front-facing video plus images (MP4, MOV, PNG, JPG, JPEG supported).
  3. 3Submit and wait for cloning; the new avatar appears in your avatar list.
  4. 4Same for voice: click "Create" next to "Select Voice" and upload a clean voice sample recorded in a quiet room.
Tip

For avatar cloning: good lighting, face the camera directly, natural expression, nothing covering the face. For voice cloning: clear articulation, no background noise.

Generate a Presenter Video

With the avatar and voice ready, one click produces the video.

  1. 1Choose the drive mode (text / audio).
  2. 2Text drive: pick the avatar and voice, then enter the script.
  3. 3Audio drive: pick the avatar and upload the audio file.
  4. 4Enter a work title so you can find it later in Works History.
  5. 5Toggle subtitles on if needed and customize the style, then click "Generate Video".
Select avatar
Choose one of your cloned avatars to appear on camera.
Select voice
Required for text drive — pick from your cloned voices.
Script
The text to be spoken in text-drive mode; its length determines the video duration.
Upload audio
The voice source for audio drive: MP3/M4A/WAV, up to 20MB, 5 seconds to 3 minutes.
Work title
Name the video so it's easy to spot in Works History.
Subtitles
When on, subtitles are generated from the speech. Open "Custom subtitle style" to adjust font, size, text color, stroke color and position (X/Y offset).

FAQ