How to create personalized AI video avatars with Veo 3
Create custom AI-generated videos featuring yourself or any character by training an AI model with personal images and then animating them with natural speech using Google Veo 3.
Required tools
None required
Updated
Jan 30, 2026
Resource
This guide has resources.
The Rundown: Create custom AI-generated videos featuring yourself or any character by training an AI model with personal images and then animating them with natural speech using Google Veo 3.
🧰 Who is this useful for:
- Content creators building personal brand videos
- Marketing professionals creating spokesperson content
- Social media managers producing consistent character-based content
- Business owners developing promotional videos without hiring actors
STEP 1: Train Your Custom Character
Head over to Freepik and navigate to the “Create” section. Look for the “Character” tab where you can create a completely new character by clicking the plus button.
Assign a name to your character, select the appropriate gender, and upload between 12-24 high-quality images of yourself or your desired character. Use varied angles, backgrounds, lighting, and facial expressions to help the AI understand your character's unique features.

STEP 2: Generate Your Perfect Scene
Once your character is trained, you'll need to create a compelling prompt for your scene. Head to ChatGPT and use a prompt like this:
“I'm experimenting with an AI image generator and I have trained it on a specific character. Provide me with a prompt where the character is in a studio, holding a product. Do not describe the character's appearance as they will be referenced already.”
This approach makes sure you get detailed, professional prompts that focus on the scene and action rather than character description, leading to better results.

STEP 3: Generate Your Character Image
Return to Freepik and select your newly created character from the library. Adjust the character strength to maximum for the best resemblance. Paste your detailed prompt, ensuring your character's name appears in the description (e.g., “@Alvaro is holding a cologne bottle in a professional studio”).
Select either Flux or Mystic as your generation model, choose a 16:9 aspect ratio for optimal video compatibility, and click “Generate.” Once you're satisfied with the result, download your image.

STEP 4: Animate Your Character
Navigate to Google Gemini and select the “Video” option. Upload the image you just created as your starting frame by clicking “Add photo.”
Create a descriptive prompt for your video animation, such as:
“Guy talks to the camera saying ‘This new cologne called Dr. Cintas is probably the best one I've tried.’”
Be specific about the dialogue and actions you want your character to perform.
Click generate and wait a couple of minutes for processing. The result will be a professional-looking video featuring your custom character delivering your scripted message with natural lip-sync and movements.

Pro tip: Experiment with different prompts and scenarios to create a library of character videos. The more specific your dialogue prompts, the better the lip-sync and overall animation quality will be.
