We put five AI avatar video tools to the test

In partnership with

We tested out five different AI video avatar tools to see how realistically they can make an image of Santa Claus talk.

We didn't just want him to look the part; we wanted to see if he could deliver a joke with the right amount of attitude. We uploaded a photo (below) of St. Nick to D-ID, HeyGen, Hedra, Veo, and Sora and gave them all similar prompts: a monologue about Silicon Valley running out of electricity and begging to use his "Christmas magic" for their data centers.

For the first three, we were able to feed the entire script. The last two work a little differently as they are limited by video length.

Become An AI Expert In Just 5 Minutes

If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.

This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.

Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.

From the subtle movements of his hands and legs to the timing of the delivery, here is how each tool handled the challenge of animating a digital Santa.

D-ID

D-ID has been our go-to tool for making images talk for more than two years now. It does a great job with voices and sticking to a script, but we’ve noticed it falling behind in realistic movement. And this output proves that.

Hedra

Hedra came on our radar with its ability to make trendy talking baby videos earlier this year. It had great movement and stuck to the script. But the quality of Santa’s face was lacking.

HeyGen

We’ve become huge fans of HeyGen in this space over the past year, and it came through with a high-quality video that stuck to the script.

Veo

Google’s Veo isn’t made to read from a script like the other tools, but we decided to give it a shot here. The movement includes an odd leg kick, but unlike the previous options the movements of the left hand/leg didn’t mirror the movements of the right hand/leg.

Sora

Finally we gave the prompt to OpenAI’s Sora. Again, this one isn’t make necessarily for avatar videos. But it had a great delivery and nice movements. It did make some clear color changes to the image however.