OmniHuman-1.5 Avatar Creation, Powered by Wondercraft
OmniHuman-1.5 is Kuaishou's AI human synthesis model — now available in Wondercraft Studio. From a single portrait and an audio track, it generates photorealistic talking video with natural lip sync, expressive gestures, and lifelike body language. In Wondercraft, OmniHuman-1.5 connects directly to 1,000+ AI voices — from portrait to talking presenter in one seamless flow.
Try it in Wondercraft
OmniHuman-1.5
OmniHuman-1.5 transforms a still portrait into a lifelike talking presenter — driven by audio, powered by Wondercraft.
Portrait-to-Presenter
Animate any portrait photo into a realistic, talking human video.
Audio-Driven Expression
Lip sync, head movement, and gesture all driven by your voiceover.
Lifelike Micro-Expressions
Subtle facial dynamics that make AI presenters feel genuinely human.
Full Upper-Body Motion
Animate beyond the face — natural shoulder, arm, and hand gestures.
Photorealistic Output
Quality that holds up to close scrutiny on any screen.
Voice-Integrated Pipeline
Generate AI voiceover and drive the avatar — all inside Wondercraft.
Wondercraft Visual Workflows
Wondercraft transforms OmniHuman-1.5 into a complete avatar production pipeline — connecting portrait animation with AI voice, compositing, and publishing.
From Portrait to Presenter
Upload a photo, generate a voice — Wondercraft creates your talking avatar.
AI Voice Integration
Choose from 1,000+ AI voices or clone your own to drive natural lip sync.
Video Composition
Layer your avatar into scenes with backgrounds, captions, and transitions.
Collaboration Tools
Review avatar videos with your team and iterate together.
Publishing and Export
Share directly, embed, or export in any format and aspect ratio.
OmniHuman-1.5 alone vs. OmniHuman-1.5 in Wondercraft
Feature
OmniHuman-1.5 Alone
OmniHuman-1.5 in Wondercraft
Capabilities
Workflow
Control
Ideal for
Portrait animation, talking video
Generate avatars, no pipeline
Audio and portrait input
Researchers, AI developers
Photo-to-presenter production
Photo to published avatar video
Voices, timeline, Wonda direction
Brand teams, educators
Flux.1 Kontext Max Alone
Long-form multi-image editing
Image generation and scene mapping
Text and style-based guidance
Visual designers, art directors
Flux.1 Kontext Max in Wondercraft
Full multimedia storytelling with audio and timeline
End-to-end production from script to release
Visual, audio, and Wonda-assisted co-direction
Storytellers, educators, and production teams
Why Wondercraft?
Presenters without cameras.
OmniHuman-1.5 creates the face; Wondercraft adds voice, editing, and distribution.
All models, one studio.
Combine OmniHuman with Kling, Veo 3.1, and image models — all in Wondercraft.
Voice-first production.
Generate voiceover and drive your avatar in the same timeline — one workflow.
Guided by Wonda.
Your built-in AI co-director helps refine tone, pacing, and presenter delivery.
How it works
FAQs
Everything you need to know about
OmniHuman-1.5
What is OmniHuman-1.5?
Kuaishou's AI model for realistic human video — animating portraits with natural lip sync and expressive body language.
Is OmniHuman-1.5 available in Wondercraft?
Yes — create and edit avatar videos directly inside Wondercraft Studio.
What makes OmniHuman-1.5 special?
Lifelike micro-expressions and full upper-body motion from a single still image and audio track.
Can I use my own voice?
Yes. Upload your audio or generate voiceover with 1,000+ AI voices — then drive the avatar.
Is it suitable for brand spokespeople?
Absolutely. Create consistent, on-brand presenters for campaigns, training, and communication.
Animated. Voiced. Published.
From portrait to published presenter video, OmniHuman-1.5 in Wondercraft Studio gives you expressive avatar animation with the AI voice, editing, and publishing tools to complete it.
Try it in Wondercraft