Overview
Create videos of an AI avatar speaking your script. Voice, lip movement and gestures are generated natively by the video model: you choose the avatar and write the script, and there is novoiceId or lipsync model to configure. Talking avatar videos are always vertical (9:16): there is no aspectRatio, voiceId or audio to send.
Get avatar IDs from GET /v1/avatar/list.
Quick Example
projectId. Poll GET /v1/project/{projectId} or pass a webhook to get the finished video.
Complete Example
Parameters
The ids
videoStyle takes are in Get Catalog: camera-moves for cameras (default most-used-still-camera-gestures), gestures for gestures (send the id, the library’s own words), voice-presets for voice and accents for accent.
Managed teams pay 100 credits per started 15 seconds of speech, estimated from the script (about 2.5 words per second). The estimate is charged when the project is created and refunded if the project fails.Teams in bring-your-own-keys mode are not charged credits for generation, but need their own OpenRouter and Gemini keys in Settings → AI keys. Without them the call answers
402 with code: "missing_credentials" and the missingProviders list, before anything is created or charged. To retry a create call after a timeout or 5xx, send it with an Idempotency-Key: the same key never creates or charges twice.Tips for Talking Avatar Videos
Effective Script Formulas
Caption Presets
caption.preset takes any id of Get Catalog with name=caption-presets (for example wrap1, tiktok, beast, classic); the default here is tiktok.
Use Cases
- Testimonials: Authentic-looking customer testimonials
- Social Media Content: TikTok, Instagram Reels, YouTube Shorts
- Personal Messages: Video messages for customers or team
- Quick Updates: Company updates or announcements
- Educational Content: Quick tips or how-to videos
- Product Presentations: Showcase products with talking avatars
Handling the Webhook Response
Webhooks are signed: check theHooked-Signature header as shown in Webhooks.