Synthesia Launches New AI Assistant for Faster Video Creation and Express-3 Avatar Model

Reimagining the video creation workflow

When we built the first version of the AI Video Assistant, we set out to solve one clear problem: helping customers move from script to video without manual work. But as customers used it, we learned about new friction points in their workflow.

The video generation itself worked and was straightforward. The bottleneck was everything surrounding it: the back-and-forth between scripting, structuring, design, and production all happening in separate places. Here is what we’re doing to reimagine the video creation experience in a way that is useful for customers: 

Move from prompt to polished, on-brand video in one conversation

Describe what you need and Assistant plans it with you, aligns on structure and tone and builds it on-brand. For example, you can type, “I want to make an onboarding video to welcome new hires,” add your docs, and let the Assistant do all the hard work. Here’s the workflow:

• Describe your video: drop in a prompt, files, or URLs and Assistant structures and scripts it for you
• Plan before you build: align on structure, tone, and scene count through a back-and-forth conversation with Assistant, then choose your delivery style (Cinematic or Presentation) before a single scene is generated
• Get fully on-brand outputs: your brand kit auto-applies across templates, avatars, and motion graphics
• Refine through in-editor chat: adjust tone, scene count, or structure without restarting

Meet Express-3: our most lifelike avatar model

Launching today alongside the Synthesia Assistant is Express-3, our most advanced avatar model for lip sync, natural body movement and gestures. Our new avatars follow the sentiment of your script, so they can be happy, serious, neutral or friendly depending on what you've written.

• Express-3 introduces more script-aligned gestures and movement, helping avatars feel less static and more human
• Videos using Express-3 can generate up to 2x faster, reducing the wait between editing and reviewing
• Express-3 powers new creation workflows, including new stock Avatars, Style Avatars, and Avatar-to-Video through Assistant
• The model was developed and evaluated through internal head-to-head testing focused on lip sync, body motion, emotional control and overall publishability
• Express-3 is rolling out in stages, starting with new Express-3 stock avatars and expanding across eligible avatar experiences over time

We’re also launching our new Avatar Builder. Realistic human-like avatars alone sometimes aren’t enough or get repetitive, while stylized, on-brand characters can keep audiences engaged. Our Avatar Builder lets you create Style Avatars: stylized, branded avatars from a prompt and preset, and soon from reference images. Our new Style Avatars keep audiences engaged and your content visually fresh. 

Name it, pair it with a voice, and integrate it straight into Assistant. You’ll unlock benefits realistic avatars sometimes do not match: easier localization with on-brand Style Avatars that can work for sensitive topics where a human presenter sometimes won’t land. Now teams can produce personalized content at scale while building or animating a brand mascot that audiences recognize and remember.

Available today

Synthesia Assistant and the full range of Express-3 Avatars are available now for all users on all plans. Learn more here.