Anyone using Synthesia for content creation?

I'm exploring AI avatars for my business content due to some practical challenges. I had an accident that limits how often I can film, and I find that my facial expressions (ASD) can be distracting in talking-head videos, especially short-form content where every second counts. While I value authenticity, I also need to deliver professional, focused content that serves my audience without unnecessary distractions.

I'm considering Synthesia as a solution, but have some concerns. For those who use it: How natural do the avatars actually look in practice? I've read they don't blink - is that as noticeable as it sounds?

I already tried training my own LoRa model. The still images turned out surprisingly well, but video was a different beast entirely. Quality, consistency, and video length all became major hurdles. After months of "I'll figure it out eventually," I'm ready to invest in a working solution now as I continue experimenting.

Would love to hear from anyone who's made Synthesia work for their business, especially if you're using it for educational content.

Alternatively, I am open to better suggestions.

2 comments