Loading…
Frame talking-head video at the source: locked 16:9 crops, brand-colored letterboxes and studio-quality framing for course and YouTube recording.
Every talking-head video inherits its webcam framing. Fix it in the edit and you're re-cropping, re-keying and re-framing every single video forever; fix it at the camera and it's done once.
CamTuner frames the feed before your recorder captures it — so course videos, YouTube segments and screencast face-cams all start with the composition you chose, not the one your webcam defaulted to.
A bad source framing means cropping and correcting in post for every upload — hours of repeated work.
Manual setups drift: this week's crop is a little different from last week's, and viewers feel it.
Screencasts need the face-cam small but well-composed — impossible with a wide, unbudgeted camera frame.
Start with a 16:9 crop as the foundation, then layer in the features that fit how content creators work:
Locked, repeatable framing
The same crop and alignment every session — your series looks like one series.
Letterbox color as a brand element
Pad your feed with any bar color to match your thumbnail palette or course template.
Record-ready for browser recorders
Riverside and other in-browser recorders capture the tuned feed — framing, blur and all.
CamTuner only adjusts camera streams, so your screen capture is untouched — but the face-cam feed inside a browser recorder or call is framed exactly as you set it.
Yes — good hardware doesn't choose your composition. The crop, alignment and consistency layer is exactly what a camera can't do for itself.
Settings live in Chrome extension storage per machine. Set the same values on your recording rig once and they persist there.
Set it up once — every call after that is already framed. No account, fully on-device.