What you'll accomplish
By the end of this guide, you'll have created a professional training video using an AI presenter avatar — no camera, no studio, no re-shoots needed. You'll go from a written script to a published video you can embed in your LMS, and you'll know how to update it in minutes whenever the content changes.
What you'll need
- A free Synthesia account at synthesia.io (free trial: 3 minutes of video)
- A written script for your training content (plain text is fine)
- Time needed: 45–60 minutes for your first video; 20–30 minutes for subsequent ones
- Cost: Free trial (limited) / $29/mo Starter (includes ~10 videos/month)
How-To Guide: Create AI Presenter Training Videos
Step 1: Create your script
Before opening Synthesia, write your video script as plain narration text — not bullet points. Write it the way you'd speak it aloud during a training session.
- Open a Word doc or Google Doc
- Write your narration as full sentences, spoken in second person ("You'll notice that...")
- Aim for 130–150 words per minute of video time (a 5-minute video = ~650–750 words)
- Add
[PAUSE]annotations where you want a brief pause (works in some Synthesia voices)
Tip: If your script feels stiff, paste it into ChatGPT and ask: "Rewrite this as natural spoken narration — conversational but professional, as if explaining to a new employee."
What you should see: A clean script with natural sentence flow, no bullet points, estimated reading time noted.
Troubleshooting: If you're not sure how long your script runs, paste it into a text-to-speech tool and time it.