Give Velo a still image, a short reference video, and your voice. Get an avatar that moves and speaks like you, reusable across every future video.
The problem was never the voice. It was a face that moved like nobody in particular.
Audio alone can't tell a model how you actually move. Avatar generation v2 gives it a reference to work from instead.
Invent the motion from audio
Same exaggerated head movement and generic lip sync for every user, because the model never saw how you actually talk.
Record yourself, every single time
Looks like you because it is you. Also means a camera setup for every prospect, every language, every variant.
Show it one clip. It remembers how you move
A still image, a short reference video, and a voice sample, captured once, generate avatars that move like you for as long as you need them.
No motion-capture rig, no repeated camera time, and no generic head movement standing in for you.
A still image, a short reference video, and a voice sample is the entire setup.
Full-frame for an intro or outro, or layered on top of a screen demo.
Swap the name, the opening line, or the avatar itself for a new version.
Edit a line in the script and regenerate just that segment, not the whole video.
1 reference clip
Is the entire setup for a reusable, personalized avatar.
~40% faster
To generate than the previous avatar model.
0 reshoots
Needed to personalize a name, a line, or a language.
A mid-market sales team wanted every high-value prospect to get a reply that felt personally recorded, but individually filming each one didn't scale past a handful of deals a week. After avatar generation v2, reps set up one avatar each and personalized the name and opening line per prospect, replying in minutes instead of scheduling camera time.
Read more →I used to skip the personal video reply whenever my calendar got tight. Now it takes less time to send one than it does to write the email it's replacing.
Free
For anyone with a story to tell, one video at a time.
$0
per month
Pro
For creators, sellers, and marketers making video part of the weekly routine.
$41
per month, billed yearly
Ultra
For teams and power users turning video into a growth engine.
$167
per month, billed yearly
Enterprise
For organizations bringing video to every team, workflow, and market.
Everything in Ultra, plus
Set up an avatar once from a reference clip, then personalize it endlessly without touching a camera again.
No. A short clip from a phone or laptop camera is enough.
Every avatar you generate now uses the new model automatically. There's no separate opt-in required.
Yes. Set it up once and place it in any number of future videos, personalizing the name, opening line, or language each time.
Edit just the part that changed and regenerate that narration and avatar segment. The rest of the video stays untouched.