avatar-video

Call POST /v2/video/generate with a video_inputs array specifying the avatar_id, voice_id, input_text script, and background per scene. Then poll GET /v2/videos/{video_id} until the status is completed to get the download URL.

Fetch the avatar's details via GET /v2/avatar/{avatar_id}/details and use its default_voice_id, which is pre-matched for natural lip sync. If no default exists, list voices with GET /v2/voices and match the voice gender to the avatar.

Yes. Upload the image to upload.heygen.com/v1/asset to get an image_key, create a photo avatar group with POST /v2/photo_avatar/avatar_group/create, then use the returned id as talking_photo_id in video generation. Avatar IV offers higher quality via /v2/video/av4/generate.

Yes, the /v1/video.webm endpoint produces WebM videos with transparent backgrounds, supporting normal and closeUp avatar styles. These can be layered over other content in Remotion using OffthreadVideo with the transparent prop.

Video generation typically takes 5-15 minutes or longer depending on length and resolution. Poll the status endpoint with generous timeouts, and use test mode during development to avoid consuming credits on watermarked drafts.

Call GET /v2/user/remaining_quota with your X-Api-Key header to retrieve remaining_quota and used_quota. Check this before generating videos, since 1080p output consumes roughly 1.5x the base credit rate.