CSA Farm Weekly Box Update With an AI Avatar: Script and Cost

A CSA farm can tell members what is in this week's box with a short host video instead of a long email. Sume Avatar 1.0 renders it in 4 to 60 seconds, with the same avatar every week.

Prices are from Sume's API pricing page and the request shape from Generate avatar video, both read on 2026-10-05. The avatar must exist first: Create new avatar accepts a prompt, a profile or a photo.

What should the script and scenes look like?

List the headline items, one tip, and pickup details. Counts use 2.8 words per second, rounded up per scene.

The seconds column is Sume's planning estimate of ceil(words / 2.8) per scene, not a measured render time.

POST https://api.sume.com/v1/avatar-1.0/talking-video
{
  "avatar_handle": "front_desk_host",
  "aspect_ratio": "9:16",
  "quality": "standard",
  "video_inputs": [
    { "id": "box", "voice": { "type": "text", "script": "This week's box has kale, carrots, red onions and the first sweet corn of the season." }, "background": { "type": "prompt", "prompt": "Bright, plain room, locked-off camera, soft daylight" } },
    { "id": "tip", "voice": { "type": "text", "script": "Roast the carrots with the onions, and save the kale stems for soup." }, "background": { "type": "prompt", "prompt": "Bright, plain room, locked-off camera, soft daylight" } },
    { "id": "pick", "voice": { "type": "text", "script": "Pickup is Thursday from three to six at the barn." }, "background": { "type": "prompt", "prompt": "Bright, plain room, locked-off camera, soft daylight" } }
  ]
}

What does it cost?

The plan above comes to 15 seconds, inside the 4 to 60 second window. At Sume's per-second rates (no product image), one render costs $2.76 on standard, $3.68 on plus (the default when quality is omitted) and $8.25 on max. Across 26 weekly updates, that is $71.76, $95.55 and $214.50. Creating the avatar is a separate one-time $0.95; later clips reuse the handle.

What if the box contents change?

Then the clip is wrong. A render is fixed once made, so confirm the contents before you render. At standard rates a correction costs only the clip's seconds again.

Keep the same avatar and scene each week so members recognise the host.

What will it not do?

It will not show the actual vegetables or answer member questions live. It is a scripted presenter.

Scripts for Avatar Video should be English; Sume's guidance is that non-English speech breaks caption alignment.