Avatars
New Feature
Update
Meet Express-3, our most lifelike avatar yet
Our best avatar model to date — sharper, faster, and more expressive, live for every user on every plan.
See what Synthesia can do for you
See all updates
Avatars are the face of every Synthesia video, so we rebuilt the model behind them. Express-3 is our next-generation avatar rendering technology, and it's replacing Express-2 across the platform.
What's new
- Avatars respond to your script's sentiment: performances shift between happy, serious, neutral, and friendly, depending on the context of what's being said.
- Sharper visuals and better lip sync: faces render more clearly, and mouth movements track speech more precisely.
- More natural body motion: gestures are more varied and better aligned with the script, with far fewer eye-gaze issues.
- Faster video generation: get your videos rendered up to twice as fast.
- New: build your own avatar: use Avatar Builder to create a custom avatar — realistic or stylized — from a starter concept, panel controls, or a text prompt, then create videos in that avatar's style.
Express-1 avatars aren't affected by this change. For more on building a custom avatar with this technology, check out our documentation.
More updates
See all updates
New Feature
Introducing Syren: our breakthrough in AI video
Describe the video, attach links or docs, and Syren directs the whole thing. Then refine it by chatting.
10/9/26
Get the skills, insights, and support to help you scale
