Thanksgiving recipe video: step cards with caption cues

Make a 40-second Thanksgiving recipe video from five phone shots: join them in Timeline 1.0, burn numbered step cards with caption cues, add a music bed.

4 min readSume
All posts

Thanksgiving is Thursday 26 November 2026. A recipe video works as five silent shots with a text card per step: join the shots in one Timeline 1.0 render ($0.10), then burn numbered step cards with cues in one video-captions job ($0.20). Total $0.30, plus a $0.125 music bed if you want one.

Layout

Example timing for five 8-second shots, 40 seconds total.
StepShotCard textStartEnd
1Chop1. Chop the onions0.56.5
2Mix2. Mix the stuffing8.514.5
3Bake3. Bake 45 minutes16.522.5
4Glaze4. Brush the glaze24.530.5
5Serve5. Serve warm32.538.5

Request

Render the joined video first, then send its artifact URL with the cues. The shots carry no speech, so cues are required; without them the job fails with caption_no_speech.

{
  "video_url": "https://media.sume.com/artifacts/artf_demo/recipe.mp4",
  "style": "tiktok-green",
  "cues": [
    { "text": "1. Chop the onions", "start": 0.5, "end": 6.5 },
    { "text": "2. Mix the stuffing", "start": 8.5, "end": 14.5 }
  ]
}

Limits

Quantities and oven times belong in your caption or description, written by you; do not let a model invent them. Cues are plain text, so a long ingredient list needs a pinned comment instead. Check the style options: the design override does not apply to tiktok-green or punch.

Sources

More in Use cases

All Use cases posts

Written by Sume