01
Reference
Choose a clear face reference. LivePersona prepares it before billable startup.
LIVEPERSONA · REALTIME AI PRESENCE
Become the persona you need — live.
Realtime FaceSwap and body-aware video transformation for live calls, with optional Sara or Malik neural voice, a clean 16:9 output for OBS, and accountable usage from one secure LivePersona account.
Realtime processed videoOptional neural voiceClean OBS output
PLANStarter
TIME REMAINING8m 20s
VOICEAvailable
LIVE CAMERARealtime FaceSwap v3Camera tested · reference ready
CAMERA READY
FACE REFERENCEReference readyChange image
BACKGROUNDOriginalOptional replacement
Sara ⌄
Uncheck to run video-only and bypass RunPod voice compute.1280×720Clean live output target
Per callNeural voice is optional
OBS readyDedicated capture surface
ReceiptsActual session usage recorded
Face & body-aware presenceReference-driven transformation designed for continuous live movement.
Voice is a choice, not a dependencyLeave it off for video-only calls. Turn it on only when Sara or Malik is needed.
Production-safe usageCredits, reservations, session receipts and provider cost remain reconciled.
FROM REFERENCE TO LIVE PRESENCE
Upload the face. Keep the motion. Send the transformed output.
LivePersona keeps the call surface compact while the cloud path handles transformation and optional neural voice behind it.
02
Move naturally
Your camera remains the motion source while the live transformation follows the session.
16:9LIVE OUTPUT03
Clean output
A dedicated capture surface stays free of diagnostics and controls for OBS and calling apps.
WHAT LIVEPERSONA DOES
One live production surface. Six controlled layers.
Each layer is deliberate. Video remains usable without neural voice, and optional features do not get to become single points of failure.
Realtime FaceSwap
Use a clear reference image to drive a transformed face while your camera remains the motion source.
Body-aware continuity
Keep head, posture and movement coherent through a continuous live session instead of isolated generated frames.
Sara / Malik voice
Enable American neural voice only when needed. Leave it off and LivePersona bypasses voice GPU compute entirely.
Background layer
Keep the original scene or introduce an optional background without turning the output surface into a control panel.
Clean OBS output
Capture a dedicated 16:9 window with no diagnostics, buttons or billing overlays on top of your video.
Usage truth
Plan credits, reservations and settled seconds remain visible instead of disappearing into opaque compute charges.
HOW IT WORKS
From reference image to a live call in four decisions.
The workflow is intentionally short because the desktop is built for the call itself, not for configuring infrastructure.
Add your reference
Choose the face image you want LivePersona to prepare.
Test the camera
Confirm framing, camera access and the local preview before billable startup.
Choose voice or video-only
Tick neural voice for Sara or Malik, or leave it off to bypass RunPod voice compute.
Run Call
LivePersona opens the processed video path and the clean output surface for your calling setup.
VOICE WHEN YOU WANT IT
One checkbox decides whether the GPU voice path exists.
Accounts with neural voice access can choose on every call. Unchecked means original/video-only operation with no RunPod voice provisioning or voice reservation. Checked reveals Sara and Malik and starts the protected voice path only when Run Call begins.
Video onlyNo neural GPU · no voice usage charge
✓
Add neural voiceSara / Malik · protected single-GPU route
Voice add-on starts at $5$5 through the first hour of plan capacity, then $3 per additional started hour. Actual converted usage remains itemized in receipts.
BUILT FOR LIVE USE
Designed around the moments where generated video has to keep moving.
LivePersona is not a one-click photo editor. Its product decisions are shaped around continuous sessions, recoverable media and a clean external output.
LIVE CREATOR
Creator sessions
Keep a distinct on-camera persona across streaming, presentation and collaborative sessions.
VIRTUAL PRESENCE
Virtual presenters
Use reference-driven presence while preserving your own live timing and body movement.
OBS / VIDEO CALL
Production workflows
Route the dedicated output into OBS and compatible call setups without exposing controls to viewers.
WHY LIVEPERSONA
The video path should survive optional features.
That principle shapes the product: voice can fail without taking video down, transient RTC interruptions get a recovery window, and account balances reconcile against actual settled sessions.
Video-first continuityOptional voice never needs to be the reason a video-only call cannot proceed.
Desktop-first controlThe controls stay compact while the output window stays clean.
Recoverable mediaShort transport interruptions are treated as recovery events before they become call-ending failures.
Measured billingCredits and provider usage are reconciled after the call rather than guessed from what a screen last displayed.
PLANS
Start with the time you need. Add voice only if you need it.
Starter
$10.008 min estimated video time · 30 days
Neural voice add-on: +$5.00
Choose StarterPro
$100.001h 23m estimated video time · 30 days
Neural voice add-on: +$8.00
Choose ProQUESTIONS BEFORE YOU RUN A CALL
Know what is optional, what is local and what gets billed.
Do I have to use neural voice?
No. Neural voice is a per-call choice. Leave it unchecked and LivePersona runs the video path without provisioning RunPod voice compute.
Do I need an NVIDIA GPU in my own PC?
The heavy live transformation and optional neural voice paths are handled by configured cloud providers. Your Windows PC still needs a reliable camera, microphone and enough resources for the desktop capture workflow.
What happens if voice capacity is unavailable?
The correct fallback is video-only continuity. A voice-capacity problem should not force the processed video session to fail.
Can I use the output with OBS?
Yes. LivePersona exposes a dedicated clean 16:9 output surface intended for capture without diagnostic overlays.
How are credits calculated?
Live sessions reserve credits up front and settle against actual usage. Neural voice access is a separate plan add-on and converted usage remains itemized.
Why does the desktop ask me to test the camera first?
It verifies the local capture path before the billable live session starts, reducing avoidable provider sessions caused by camera permission or device issues.
LIVEPERSONA FOR WINDOWS