Avatar flash-sale video in three scenes: hook, silent demo beat, CTA

One request, three beats

Use video_inputs on POST /v1/avatar-1.0/talking-video and give each beat its own entry: a spoken hook, a silent product moment, and a spoken call to action. A silent beat is voice.type: "silence" with a required duration, and it cannot carry script or input_text.

Black Friday is Friday, November 27, 2026 and Cyber Monday is Monday, November 30, so a flash-sale clip can name the window directly. Provide exactly one of script or video_inputs; the planned total must land between 4 and 60 seconds.

The request

Replace sume_clawra with the handle of a ready avatar of your own. Scenes are expected to resolve to one shared scene, and one resolved avatar per final video, so keep the same background prompt in each scene.

curl -sS -X POST https://api.sume.com/v1/avatar-1.0/talking-video \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: bf-flash-001" \
  -d '{
    "avatar_handle": "sume_clawra",
    "aspect_ratio": "9:16",
    "quality": "plus",
    "captions": { "enabled": true, "style": "slam", "language": "auto" },
    "video_inputs": [
      { "id": "hook", "voice": { "type": "text", "script": "Friday only: the whole shop is 30 percent off.", "duration": 4 },
        "background": { "type": "prompt", "prompt": "Bright home studio, soft window light" } },
      { "id": "demo", "voice": { "type": "silence", "duration": 3 },
        "background": { "type": "prompt", "prompt": "Bright home studio, soft window light" } },
      { "id": "cta", "voice": { "type": "text", "script": "Code FRIDAY30 at checkout. Ends Monday.", "duration": 4 },
        "background": { "type": "prompt", "prompt": "Bright home studio, soft window light" } }
    ]
  }'

Quality and ratio choices

Pick the tier by how many variants you need. Resolution is currently 720p.

Check the placement and the captions

Inline captions burn into the final MP4 from the spoken text, and a caption failure soft-fails: the job can still succeed with a clean video_url and captions.status=failed. Inline captions are not a separate billed caption job.

For TikTok in-feed placement the ads page lists vertical 9:16 non-Spark video at 540 x 960 px or larger, so the 9:16 default fits; compare the length of your plan to the page before you buy a placement (read 2026-10-03).