A video script for AI video generation should do more than list narration. It should define the scene, visual subject, motion, timing, brand style, source image, voiceover, captions, and review criteria.
This template works for a video tutorial script, a short product video, an AV script, or a branded image-to-video workflow.
For production, use AI Image to Video, AI Video Generator, or the conversational Video Agent.
Quick Answer
To write a video script example for AI generation, split the video into short scenes. For each scene, write the narration, visual action, source image or subject, motion direction, duration, brand rules, text/caption needs, and quality checks. Then convert each scene into an AI video prompt.
AI Video Script Template
Project: [video name]
Goal: [teach, launch, explain, compare, onboard]
Audience: [who will watch]
Channel: [YouTube, landing page, TikTok, product page, ad]
Format: [16:9, 9:16, 1:1]
Length: [15s, 30s, 60s]
Brand style: [palette, lighting, mood, pacing]
Primary CTA: [next step]
Scene 1:
- Purpose: [what this scene explains]
- Duration: [seconds]
- Narration: [spoken line or caption]
- Visual subject: [product, person, interface, diagram, environment]
- Source image: [reference image or generated still]
- Motion: [camera push, pan, product rotation, hand movement]
- Prompt: [AI video prompt]
- SEO/publishing note: [thumbnail, filename, caption, transcript line]
- Review checks: [accuracy, crop, brand fit, no fake text]
This structure keeps the video script connected to the generated output. A plain paragraph of narration is not enough for an AI video workflow.
Video Tutorial Script Example
Example topic: how to refill a skincare product.
| Scene | Narration | Visual direction | AI video prompt |
|---|---|---|---|
| Hook | "Refill your moisturizer in under a minute." | Product and refill pouch on clean counter | Close-up product setup on cream counter, natural light, slow camera push, clean space for caption |
| Step 1 | "Open the outer jar." | Hand removes lid | Hands opening refillable skincare jar, gentle motion, accurate product shape, no readable text |
| Step 2 | "Insert the refill pod." | Refill pod slides into jar | Refill pod placed into jar, smooth motion, minimal background, brand palette |
| Finish | "Replace the lid and store it normally." | Final product hero shot | Finished moisturizer on bathroom shelf, soft shadow, calm premium mood |
Use Image to Video AI Guide when you already have a source image and need motion prompts.
AV Script Template
An AV script separates audio and visual direction.
| Time | Audio | Visual | Notes |
|---|---|---|---|
| 0-3s | "Meet the refill system." | Product hero, slow push-in | Keep label area clean |
| 3-8s | "Swap the pod, not the whole package." | Refill sequence | Show hands, avoid fake text |
| 8-13s | "A simpler routine with less waste." | Product in bathroom scene | No quantified claim unless approved |
| 13-15s | "Shop the refill bundle." | Packshot with CTA space | Add CTA in editor, not generated text |
For AI generation, the visual column becomes the prompt base. The audio column becomes voiceover, caption, or landing page copy.
Prompt Framework for Each Scene
Create a [duration] AI video scene for [brand/product].
Scene purpose: [hook, step, proof, transition, CTA].
Visual subject: [source image, product, person, environment].
Motion: [camera movement, object movement, pace].
Brand style: [palette, lighting, mood, composition].
Channel: [16:9 tutorial, 9:16 short, square ad].
Text handling: leave space for editable captions, no generated readable text.
Quality controls: accurate product, stable geometry, natural motion, clean crop.
If the video starts from an image, generate or choose the still first, then write motion around it. This gives the model a clearer visual anchor.
Technical SEO for Published Videos
When the video is embedded in a blog post, tutorial, or product page, include crawlable context around it.
- Add a descriptive page heading near the video.
- Include a short transcript or step summary in text.
- Use a descriptive video filename when self-hosting.
- Create a custom thumbnail with an accurate alt text or image context.
- Add captions when possible.
- Place the video near the section it supports.
- Use VideoObject schema only when the page has the required visible details.
- Link to related guides such as Sketch to Video AI Guide and Veo 3 JSON Prompt Generator Guide.
Review Checklist
Before publishing an AI-generated video:
- The script matches the page or campaign goal.
- Each scene has one visual job.
- The product is accurate.
- Motion is natural and not distracting.
- Text is added in editing when readability matters.
- Brand style is consistent across scenes.
- Captions or transcript support accessibility.
- Claims are supported by verified facts.
FAQ
What is a video tutorial script?
A video tutorial script is a scene-by-scene plan that explains what the viewer hears, sees, and does. For AI video generation, it should also include visual subject, motion, duration, brand rules, and prompt direction.
What is an AV script template?
An AV script template separates audio and visual directions in columns. It is useful when narration, on-screen action, captions, and review notes need to stay aligned.
How do I turn a script into an AI video prompt?
Break the script into scenes, identify the visual subject for each scene, add motion and brand constraints, then write one prompt per scene. Keep each prompt focused on one action.
Should AI video prompts include spoken text?
Include narration for planning, but avoid relying on generated readable text inside the video. Add captions, labels, and CTA text in editing when accuracy matters.
Can one image become a full video?
Yes, an image can become a short scene with camera motion or subtle subject movement. Longer videos usually need multiple stills or scene prompts.
Does embedding a video help SEO?
It can help user experience when the video supports the page, but it should be paired with crawlable text, clear headings, captions or transcript, and relevant internal links.