Pick a character + driving input, select a provider, review estimate, and queue the render.
Avatar speaks a script OR transfers a performance from a reference video.
Character photo + script → lip-synced talking avatar (HeyGen / Creatify).
Character + driving video → full mimicry, lip-sync, expression (ElevenLabs, ≤10s).
Character + driving video → movement + camera path only (ElevenLabs, ≤30s).