Generate vivid, personalized videos featuring multiple scenes — each with its own avatar, voice, background, and styling — using a structured list of inputs.
🚀 Follow our Quickstart Guide below to get up and running in minutes.
avatar_id
and a voice_id
for each speaking segment.
avatar_id
voice_id
voice_id
from the accent you want in your request. This provides fine-grained control over tone, language, and personality.
voice_id
to override the default voice associated with an avatar.voice_id
is not provided, the system will use the default voice linked to the selected avatar.
💡 Each voice includes multiple accents — select the one that fits your content best using the provided id
.
GET /api/voices/
id
from the accents
list as your voice_id
in the video generation request.
⚠️ Save the id
— you’ll use it to check progress.
⏳ Once thestatus
isdone
, download the video from theoutput
field.
Step | Endpoint |
---|---|
Get Avatars | GET /api/personas/ |
Get Voices | GET /api/voices/ |
Create Task | POST /api/lipsyncs/multi_avatar/ |
Check Status | GET /api/lipsyncs/{id}/ |
API Reference | Multi-Avatar Reference |