Training knowledge-check clips with an AI avatar, one per question

For a training knowledge check, render two short clips per question with the same avatar: a 10-second question and a 10-second answer. On Sume standard that is $1.84 per clip, so 20 questions as 40 clips cost $73.60; on plus it is $98.00 and on max $220.00. Sume renders the video only: scoring, branching and pass marks belong in your learning system. See the avatar video docs.

Use a silence beat for thinking time

In one video_inputs plan a scene with voice.type silence and a duration makes the avatar pause without speech. A question scene, a silence scene and a prompt scene fit in one clip if the total stays within 4-60 seconds.

• Spoken scenes use script or input_text, not both.
• Silence scenes need duration and no text.
• All scenes share one avatar and one resolved background.

Cost for 20 questions

Forty 10-second clips, no product image.

One question clip

Keep a file mapping question id to job id.

curl -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: quiz-q01-question" \
  -d '{
    "avatar_handle": "trainer",
    "aspect_ratio": "16:9",
    "quality": "standard",
    "script": "Question one. A customer reports a failed export. What is the first thing you check? Pause and decide, then continue to the answer clip."
  }'

curl https://api.sume.com/v1/jobs/JOB_ID/status -H "Authorization: Bearer $SUME_API_KEY"
curl https://api.sume.com/v1/jobs/JOB_ID/result -H "Authorization: Bearer $SUME_API_KEY"