Make videos without recording yourself 🎥 | Inbox Experiments #2

Welcome to another solid issue of Inbox Experiments 👋

You want to create video content; but don’t want to film yourself.
Maybe you’re camera shy, or you are an introvert. Or maybe you just don’t want to spend 30 minutes getting the lighting right.

This week, I ran a few tests to answer one question:

Can you create realistic, personalized videos without ever pressing "record"?

Turns out, yes.
Here are the best methods I tried 👇

🧪 This Week’s Experiment

How to create video content without recording yourself

🎯 Goal: Produce talking-head style videos using AI-generated versions of your face and voice
🛠 Tools tested: HeyGen, ElevenLabs, Flux, Kling, Argil
💰 Budget: $0-45 depending on the path
⏱ Time: ~10–90 minutes

🔹 Option 1: Build your AI twin with HeyGen (most natural)

HeyGen lets you upload a short video of yourself. (minimum 2 minutes to create a good output)
It analyzes your face, expressions, and voice, and turns it into a reusable avatar.

✅ After that, you can type any script and your AI self will deliver it
✅ It even creates a voice clone from that single video
✅ It’s the most natural and emotionally expressive result I’ve seen

🎥 Tip: Use good lighting and audio, your original video becomes the foundation.

🔹 Option 2: Use a static photo in HeyGen (fast but flat)

This method lets you upload a photo and generate a talking avatar.

✅ Fast and easy
❌ But: The result feels robotic; limited gestures, stiff delivery
❌ I wouldn’t recommend it unless you're just testing the flow

🔹 Option 3: Voice-only creation with ElevenLabs

You don’t need an avatar to start, just a voice.

✅ Record 1–2 minutes of yourself speaking
✅ ElevenLabs generates a high-quality voice model
✅ You can use that voice anywhere: over slides, visuals, animations

🔊 Ideal if you’re camera-shy or dont want to show up your face but still want a recognizable personal brand

🔹 Option 4: Cinematic avatar workflow (more stylized)

Want something more creative and aesthetic?

Here’s what I tried:

1. Trained a LoRA model on my photos using Flux

2. Generated images with strong cinematic vibe

3. Animated them using Kling

4. Added my HeyGen voice on top

🎞 Great for storytelling, moody visuals, or experimental brands

💡 Add-On: Argil.ai

If your avatar clip feels too static, Argil helps bring it to life:
🖼 Add relevant B-roll
💬 Insert transitions, captions
🧠 Even convert long-form content into structured scenes

A great way to go from “good AI video” to “publishable content.”

🛠 Tool Stack Recap

• 🎥 HeyGen: Create video avatar + voice

• 🗣 ElevenLabs: High-quality voice cloning

• 🖼 Flux with Fal: Personal image training

• 🎬 Kling: Animate static images

• 📹 Argil: Add layers, transitions, B-roll

🧠 Personal Insight

This isn’t “faceless content.”
It’s you, in a new form but automated, scalable, low-friction.

You still show up. You still deliver your message.
But now, AI handles the overhead.

For me, it unlocked a new way to be present without burning energy on every take.

🌟 Community Spotlight

Shoutout to Ayush, creator of SuperFrameworks — a killer newsletter packed with actionable breakdowns for solo founders.

If you enjoy Inbox Experiments, you’ll love SuperFrameworks → Check it out

📩 Got a tool, product, or experiment worth sharing?
Hit reply and send it over and let me feature you in the next issue.

📬 Before You Go...

Would you try any of these AI-powered workflows?
Do you know a creator using similar tactics?

👉 Hit reply and tell me.
I’d love to feature creative experiments in future issues.

See you next Wednesday,
Yunus 🚀