Three steps to live
No dataset, no training run, no export queue. Load your photo and voice clip, tune the look, and go live.
Pick your source input
Select a source photo for the face swap; load several to switch on the fly, or average a few photos of one person for a more stable identity. Add a short audio clip (a few seconds is enough) as your real-time voice reference.
Dial in the look and sound
Choose your swap model (HyperSwap or inswapper) and resolution, enhance detail, restore expressions, add anime or cinematic looks, reshape or re-age the face, tune your cloned voice, then pick your camera and mic.
Go live
Hit Go Live and NZCAM streams your swapped face and cloned voice into the NZcam Camera virtual camera and virtual microphone; pick them in OBS, Zoom, Teams, Meet, Discord, or any app. Hit Stop to end the session.
Three modes
Pick the mode that fits how you're using NZCAM.
Live
Webcam → virtual camera and microphone, for streaming and calls.
Image / Video
Process a photo or video file to a file; the full realism suite and colour looks apply.
Headless / CLI
Run NZCAM without the UI for scripted or automated workflows.
Try it yourself
Subscribe to NZCAM and follow these three steps to go live.
Get NZCAMFor lawful, consensual, 18+ use