Last week, I was sick and lost my voice.

So, I figured it was the perfect time to experiment with AI avatars on my tiktok, instagram, and youtube.

FYI my avatar always wear a hat 🤠 

On tiktok, I labeled most of these videos as “AI generated”, anticipating this would throttle views. Honestly, I was nervous this experiment could PERMANENTLY hurt my social accounts.

But, I was in for a surprise!

My AI avatar videos from last week racked up 2.1M+ views in total.

Not too shabby 😅 

So, let’s walk through how to make your own AI avatar that talks on camera for you and generates short-form content that can outperform human videos.

In this guide, you’ll learn:

• How to create a realistic AI avatar

• How to script and edit videos with it

• How to choose visual backgrounds for better engagement

• How to publish consistently without filming or editing

Step 1. Film Yourself

You’ll need HeyGen to get started.

This app generates highly realistic avatar videos synced to your own custom script. It’s simple, fast, and works well for talking-head short-form videos.

Start with the Creator Plan ($29/month). This allows you to make unlimited avatar videos without background removal, which is honestly good enough. You can always upgrade later if you want background effects like green screen.

Here are some tips to record “training data” video footage for your avatar:

• Use your phone’s front camera. You don’t need a fancy $5k camera setup.

• Film yourself staring directly into the camera, talking for 5 continuous minutes. You don’t need to make sense. I didn’t know what I was saying anymore after 2 minutes of continuous talking 🤣 

• Stay still. Minimize nodding, movement, and hand gestures.

• Speak clearly, but slightly exaggerate your inflection/emotions to help the avatar sound more human.

• Sit outside in natural sunlight. Avoid harsh sun but aim for even, natural lighting. This step alone is worth $10k in fancy camera gear.

• Do not stitch together clips. One continuous video is required, else HeyGen will reject it.

Step 2. Create Your Avatar

After recording yourself for 5 minutes, upload it to HeyGen to generate your avatar:

• Go to “Avatars” in HeyGen and select “Create Avatar.”

• Choose “Hyperrealistic Avatar” and select “Still” mode (not Motion).

• Upload your recorded clip.

• Record the consent video when prompted (this allows Heygen to use your likeness).

• Your avatar will take some time to render. When it’s done, you can preview and adjust the voice settings.

You can create up to 100 different “looks” per avatar.

This is handy to mix things up, so that it’s not the same background or clothing every time. However, it’s not necessary because my avatar videos all used the exact same avatar, same clothing.

If you want the green screen effect, you’ll need to open your avatar and click “Edit”.

Then click “Background Removal” in the top right corner, but this requires upgrading to the next paid tier. Finally, click “Save as New” which will save this avatar as a background-less version.

Label these clearly so you can swap between them.

Step 3. Create Your 1st Avatar Video

Now let’s make your first talking-head AI avatar video!

• Go to HeyGen

• click “Create Video”,

• click “Avatar Video”

• click “Create Portrait Video”

• Select your avatar

• click “Script” in the left sidebar

For testing purposes, start with a simple 1 sentence script.

Click the “Play” button to preview your AI avatar voice reading the script. Every time you edit the script, you can click the “Play” button to regenerate your avatar’s voiceover.

Step 4. Export Video

Click “Preview” in the top right corner to preview your AI avatar video.

Make changes, as needed, and when you’re satisfied, hit “Submit” in the top right corner. It may take a few minutes for HeyGen to render your video.

When it’s done, you can download it, then upload to your social media.

And that’s it! 🙌 

Advanced Optimization Strategies

I’ve included advanced customization options below if you’re looking for the highest-quality avatar possible.

1. Use ElevenLabs Instant or Professional Voice Cloning

If HeyGen’s default voice profiles feel too robotic, integrate your custom voice via ElevenLabs.

• Go to “AI Voice” inside HeyGen

• Select “Integrate Third-Party Voice”

• Add your ElevenLabs API Key

• Select your ElevenLabs voice when creating avatar videos

In the HeyGen video editor, click the “More” icon next to the “Play” voice icon, and you can adjust your voice’s settings.

You might need to regenerate a few times to get a natural-sounding take.

2. Add Green Screen Backgrounds

As discussed, you’ll need to upgrade if you want HeyGen background removal.

This unlocks green screen support, which allows you to:

• Add dynamic video backgrounds

• Add photo/image backgrounds

To add images/videos to your avatar video, open the HeyGen video editor:

• Click “Assets” and upload images/screenshots or videos

• For example, add 2-3 image slides in sequence for visual engagement

I personally like to resize your avatar, make it a bit smaller, and anchor it to the bottom right corner. But it’s totally up to you.

This gives space for background visuals while still featuring your avatar onscreen.

Make sure to match your background color with the dominant tone of your media (e.g. if you have dark mode Reddit screenshots = black canvas behind avatar).

3. Automation

Once you’ve nailed a proven format, you can automate your avatar workflow via:

• Make.com or n8n (for workflows)

• Perplexity or ChatGPT (for writing your scripts)

• Blotato (for creating faceless video backgrounds & posting to all platforms)

Here’s a step-by-step tutorial and template if you’re interested in avatars + automation.

Need More Help? 👋 

1/ If you want to grow on social media and build a business in coaching, consulting, speaking, selling apps, or digital products… check out Blotato

2/ Free AI courses & playbooks here