How to use these scripts
Recording guide & conventions
Zero to AI Founder — Course Video Scripts (Google Vids edition)
Full recording scripts for every module of the Zero to AI Founder course by CGW Systems, structured scene-by-scene for Google Vids.
How these scripts are built
Every lesson is broken into numbered scenes. Each scene has exactly two parts:
- Visual — what's on screen for that scene (a slide, a screen recording, a diagram, a title card, a photo).
- Voiceover — the exact words to speak for that scene, written the way you'd actually say them. Read it aloud once before recording; if you trip, rewrite it.
That's the whole format. One scene = one visual + one voiceover. It maps directly onto how Google Vids assembles a video, so you can build each lesson straight through with no translation step.
Built for Google Vids' text-to-speech
Every Voiceover line is pre-cleaned for Google Vids' AI voiceover:
- No bracketed stage directions. Director's notes like "[live build]" or "[show it]" have been removed from the spoken text — they're cues for you, not words to read aloud, and a TTS engine would narrate them as garbage. The action they pointed at is described in plain words instead.
- No fill-in placeholders in the spoken line. Template blanks are written as the words you'd actually say (e.g. "your specific person"), and worked examples use concrete phrases a voice can read naturally.
- One name, one voice. The on-camera name is set to "Cedrick" throughout — if you're recording under a different name, find-and-replace it once before you generate the audio.
So you can paste each scene's Voiceover straight into the AI voiceover field with no editing. The only thing the TTS engine reads is the sentence you want spoken.
How to build a lesson in Google Vids
- Open Google Vids → New video → start from a blank or a template.
- One scene per number. For each scene in the script, add a new scene (the "+" / add-slide control). The scene order in the script is the scene order in the video.
- Set the visual. Use the Visual line as your guide:
- "Title slide" / "Title card" → a title layout with that text.
- "2x2 matrix" / "timeline" / "bars" / "flowchart" → build it with Vids' shapes, or drop in an image/diagram you made elsewhere.
- "Screen recording" / "live demo" → record or paste the clip; trim it to the scene.
- Keep one idea per scene. If the visual needs two ideas, split the scene.
- Add the voiceover. For each scene, paste that scene's Voiceover text straight into the AI voiceover field — it's already TTS-clean, so no editing needed. (Or record yourself reading it and drop the audio onto the scene.) Pick one voice for the whole module — consistency reads as professional.
- Time the visual to the voice. The visual should stay on screen for the full length of the voiceover. If a visual needs more time, the voiceover is probably too thin — expand it. If the voiceover outlasts the visual, split into two scenes.
- Captions on. Turn on captions for every scene. They help retention, accessibility, and people watching with the sound off.
- Tight cuts. Remove the "umms," the restarts, the dead air between scenes. A 20-minute lesson usually becomes a 14-minute lesson after a ruthless cut.
Scene layout cheat sheet
| Visual cue in the script | Google Vids layout to use |
|---|---|
| Title card / title slide | Title layout, big text, minimal |
| Single sentence / one-line takeaway | Full-bleed quote or centered text |
| 2x2 matrix / quadrant | 4-box grid with shapes |
| Timeline | Horizontal arrow with labeled stops |
| Bars / chart / curve | Chart layout or imported chart |
| Flowchart / decision tree | Shapes + connectors |
| Side-by-side (before/after, vs.) | Two-column layout |
| Screen recording / live demo | Media scene with the clip |
| Bulleted list / checklist | Bulleted text layout |
| Table / comparison | Table layout |
Recording tips (still worth reading)
- Set up once, batch a whole module. Same camera, same mic, same lighting, same background for every scene in the module. Consistency saves you hours and looks intentional.
- Audio first. A cheap lavalier or USB mic in a quiet, soft-furnished room beats an expensive camera with echoey audio. Viewers forgive bad video; they click away from bad audio.
- Record the voiceover and the visual separately. Record your voice (or paste the text into the AI voiceover) first, then build or record the visual to match its length. Faster than narrating perfectly while clicking through slides live.
- Keep the AI voice honest. If you use Google Vids' AI voiceover, listen back once. Fix any mispronounced terms by re-recording just that scene yourself, or tweak the wording so the AI says it cleanly.
- One idea per scene. If you find yourself cramming two ideas into one scene, split it. Shorter scenes keep attention; long scenes lose it.
File map
course-scripts/
README.md ← you are here
01-ai-foundations.md ← Module 1 (free preview)
02-finding-your-idea.md
03-building-your-product.md
04-go-to-market.md
05-pricing-and-monetization.md
06-scaling-and-operations.md
Each module file lists its lessons in order. Each lesson lists its scenes in order. Build top to bottom and you'll come out with a finished video per lesson.
After you record
Once each lesson video is exported, host it somewhere embeddable (YouTube unlisted, Vimeo, or Google Drive) and send me the URLs. I can build a lesson player page in the app that embeds each video and ties into the progress tracking already on your dashboard — so students mark lessons complete as they watch. Just say the word and drop in the links.