THE FULL STACK · HEYGEN
I did not film this video.
How I make AI avatar videos of myself with HeyGen, and what I would tell you before you try it.
I am a HeyGen Ambassador and partner. If you sign up through this link, I may earn a commission, at no extra cost to you.
Read the transcript
I did not film this video. I never stood on this stage. But every word you are about to hear is mine, and I made it with a tool called HeyGen.
I am Kevin Barrett. I help small and local business owners put AI to work. HeyGen lets you create a video version of yourself. You give it a photo or a short recording, and it builds an avatar with your face. You clone your voice. Then you type a script, and your avatar delivers it. No camera. No studio. No retakes.
Here is how I make mine. I create a few looks of myself: a stage shot like this one, and two close-ups in the same outfit. I write one script, split it across the scenes, and HeyGen renders it as one finished video. Once the template is set up, a new video is mostly a new script.
Now the honest part, because you deserve it. HeyGen changes its layout often. My cloned voice once beeped on certain words until I learned to write them out. And setting up your first avatar takes patience. That is why I do not learn it alone. I use AI to help me use AI, and I teach owners to do the same. One more thing you should know. The HeyGen link on this page is a partner link. If you sign up through it, I may earn a commission, at no extra cost to you. I only recommend it because I use it every week.
And yes, this is my personal AI avatar. My face. My voice. My words. If you want to try it yourself, the link is below. If you would rather build it with someone beside you, book a conversation. With you. Not for you.
What I use it for
I use HeyGen every week, and I am a HeyGen Ambassador.
- My homepage video, "What I Do: The Co-Elevation AI & HI Method" (November 2026).
- "On My Way Up," the anthem for this site, performed by an AI avatar (November 2026).
- Weekly Chronicle videos, such as "The Value Was Already There" and "The Tool Is Not the Conscience."
- This page's video, built from one reusable three-shot template: a stage shot, a medium close-up and a tight close-up, all in the same outfit.
What I learned the hard way
- Choose the voice model on purpose. My cloned voice sounded natural on ElevenLabs v4 and robotic on Eleven Multilingual v2. Set v4 as your default.
- Write contractions out. "That is" and "It is" fixed the beeps in my cloned voice.
- Play every scene before you generate. If one is silent, retype its text to force fresh audio.
- Make one look per shot, all in the same outfit, so the cuts read as one person.
- Build your avatar from a clear, real photo of yourself. An avatar made from an AI-enhanced image may fail HeyGen's identity check.
Who it is for, and who it is not for
It is for you if…
- You need to show up on video regularly, without filming.
- You will write your own scripts in your own words.
- You will review each scene before you press Generate.
It is not for you if…
- You want a finished video in one click, with no setup.
- You are comfortable on camera and can record on your phone quickly. That may serve you better.
You own the account and the tools. If you would like help setting it up in your business, that is what the Co-Elevation Method is for.
