Upload a clear front-facing portrait. The system checks clarity, occlusion, and multi-person risk.
Select a default voice and use preview to compare speaking styles before generating the avatar.
After generation, continue into the experience area and configure actions, outfit, and background.
Support real-time voice calls so users can talk to the avatar naturally during a live session.
Support long-running live generation for continuous avatar interaction without fixed duration limits.
Create a personalized avatar from uploaded photos, full-body image generation, name, and voice settings.
Freely change clothing and backgrounds to match livestream, product, classroom, or brand scenarios.