Skip to main content
Open Avatars in the sidebar. The Studio tab builds a new avatar video; the Library tab keeps your avatars and the videos you’ve made.
The Avatar Studio with the avatar picker, the script editor and the preview

① Library / Studio · ② Choose avatar, Upload photo or Create new · ③ Write script or Upload audio · ④ Preview / Examples · ⑤ Language, voice and music · ⑥ Generate video

​
1. Choose your avatar

Three ways to get a face (②):
  • Choose avatar — pick one of the curated faces, ready to use.
  • Upload photo — use your own photo. A front-facing, well-lit portrait from the shoulders up works best, with a plain background and nothing covering the face. Only upload people who agreed to it.
  • Create new — describe a character and AI generates a brand-new face: a presenter, a mascot or a stylised character.

​
2. Write the script

Choose how the avatar gets its voice (③):
  • Write script — type what the avatar should say, up to 5,000 characters, and choose the Narration Language (English, Spanish, Portuguese, French, German, Italian, Chinese, Japanese and more under See more).
  • Upload audio — use your own voice track. The avatar lip-syncs to it, so you don’t need a voice.
Write the way people talk: short sentences, one idea each. Punctuation drives the pauses — a comma is a short breath, a full stop a longer one. Read it out loud once before generating.

​
3. Pick a model and a format

The Pick a model step with HeyGen Avatar 4 selected and the aspect ratio options

① Change model · ② The selected model, its tier and price per second · ③ Quality, realism and speed · ④ Aspect ratio

HeyGen Avatar 4 is selected and recommended. Click Change (①) to compare the four models side by side:
The model picker with HeyGen Avatar 4, OmniHuman, Kling AI Avatar v2 Standard and Pro

① What the model lets you control · ② Tier · ③ Price per second of video

The cost is the length of the video times the price per second — a 30-second video with HeyGen Avatar 4 costs about 150 credits. The exact total is always shown at the bottom before you generate. Then pick the Aspect Ratio (④): 9:16 Portrait for TikTok, Reels and Shorts, 16:9 Landscape for YouTube and webinars, or 1:1 Square for Instagram feeds.

​
4. Voice, music and captions

Further down the Studio:
  • Voiceover — over 1,800 voices, with featured ones on top; Browse all opens the full library. Your cloned voices are here too. (Not needed if you chose Upload audio.)
  • Background Music — the Crealix library, My Music (tracks you made in AI Music) or Custom uploads. Planning to publish on YouTube? Some library tracks may cause copyright claims; your own uploads or music made in Crealix are the safer choice.
  • Caption Style — 48 presets in the Crealix Library, plus your own styles under My Captions.

​
5. Performance controls

Open Performance controls at the bottom of the Studio to direct the delivery.
Performance controls: quality profile, performance intensity and facial expression

① Quality profile · ② Performance intensity · ③ Facial expression

  • Quality profile (①) — 360p Fast draft, 480p Standard, 540p HD, 720p Creator quality (default) or 1080p Cinematic. Test a script at a low resolution, then make the final at 720p or 1080p.
  • Performance intensity (②) — Natural & composed for a calm presenter, or Animated & expressive for more energy and movement.
  • Facial expression (③) — for example Happy expression for a friendlier delivery.
The controls depend on the model — these are HeyGen Avatar 4’s. The Kling models show Style direction instead.

​
6. Preview and generate

Preview (④) shows your avatar in the chosen format; Examples plays sample videos. Below it you see a summary of the language, voice and music (⑤). Click Generate video (⑥) — credits are only charged when generation starts, and failed generations are refunded. Finished videos appear in the Library tab and in your media library.

​
Use avatars in other videos

Avatar videos are saved to your media library. You can:
  • Place one as a video overlay on top of any Crealix video — like a presenter in the corner. See Text overlays & your logo.
  • Use one as a source in a Split Video — for example a presenter on top and gameplay below.
  • Upscale it to 2K or 4K.