The Problem
Content at scale needs hundreds of animated typography and infographic assets every week. Making each one by hand in After Effects, keyframe by keyframe, turned motion graphics into the bottleneck of the whole production.
| Step by hand | Why it doesn't scale |
|---|---|
| Listen and mark where each word lands | Repeated for every line of every video |
| Decide which words to emphasise | A judgement call, made hundreds of times a week |
| Keyframe the type to match | Slow, fiddly, and easy to get out of sync |
| Export and import into the edit | Another handoff per asset |
How It Works
AutoEdit automates the loop end to end. Whisper aligns every spoken word to its exact time, Gemini reads the meaning and picks the keywords worth highlighting, and the studio renders the animation as SVG and Canvas at 60 frames per second.
Key Decision: Split Timing From Meaning
The two hard parts of kinetic typography are different problems, so each gets the tool that's good at it.
| Problem | Tool | Why |
|---|---|---|
| When is each word said? | Whisper | Word-level timing from the audio itself, so type lands on the beat |
| Which words matter? | Gemini | A semantic read of the script, so emphasis follows meaning |
| How should it move? | The studio | SVG and Canvas animation, consistent across every asset |
The Result
An editor drops in a voiceover and gets animated typography back, timed to the speech, ready to export into their video editor in one click.
| Before | After |
|---|---|
| Keyframing every asset in After Effects | Generated from the voiceover |
| Emphasis chosen word by word | Keywords picked automatically |
| Motion graphics as the bottleneck | Motion graphics keep pace with the edit |
AutoEdit is an internal tool, so there's no public build or source to link to.
