Pass the scene.
Steal the voice.
Load a finished pack, gather everyone around the projector, then work through the marked lines one at a time. EchoCut Studio stays one click away whenever the timing needs fixing.
Quick-launch a file into EchoCut Studio
Drop a video, .dbpack, ZIP, or RAR here when you need to edit rather than start a room.
VoxRelay Converter
Make oversized or incompatible video packs smaller and easier to run on Chromebook, laptop, desktop, iPhone, iPad, and Android — entirely on this device.
Smaller file. Same scene.
Checking this device…
Choose any media file or finished dub pack
Any local media file or VoxRelay/ChoicerVoicer pack. VoxRelay probes it first, then chooses the fastest compatible free engine.
Converted file ready
Single player
Import one finished ZIP, DBPACK, or RAR pack and start reenacting. No lobby or room code.
Choose your dub pack
Load the pack, make sure it has marked lines, then go straight into the reenactment.
Drop a finished ZIP / DBPACK / RAR pack here
ZIP, DBPACK, and RAR packs load their video and marked dialogue lines locally.
Session lobby
Load the pack, preview the scene, then start when everyone is ready.
Drop a finished ZIP / DBPACK / RAR pack here
The original pack stays intact. VoxRelay reads its bundled video for the projector and its marked dialogue for the round.
Projector waiting for a pack
Load a finished .zip, .dbpack, or .rar pack to start the preview.Session tools
Useful controls before everybody starts recording.
Multiplayer is live: the host's ZIP/session content, projector playback, player ready states, cast choices, Co-op/PvP mode, and round settings are synchronized with everyone connected to this room. Single player remains fully local/offline.
Join a session
Enter the room code and the name you want shown in the lobby.
Enter the host's room code. VoxRelay can connect across different Wi-Fi, cellular networks, cities, and regions using direct WebRTC plus a secure WebSocket internet-bridge fallback.
Perform the lines
Hear the cue, record your take, review it, then move on.
The reveal
Your recorded lines are assembled into one finished reenactment video.
No take selected
—
PvP showcase
Every competitor dubbed the complete scene. The host controls whose replay appears next.
Player
Complete-scene reenactment
Player
Full-scene PvP reenactment
The full PvP showcase has been viewed. Download your own compiled MP4 or every player's replay as one ZIP.
The Verdict
PvP voice-match results for the complete scene.
And the winner is…
WavLM Base+ neural speaker embeddings are fused with timing, rhythm, pitch contour and vocal tone. Missing blue-waveform clips reduce the final score.
PvP Results uses a quantized WavLM Base+ speaker-verification model in the browser, fused with VoxRelay’s timing/rhythm/pitch/tone judge. The model downloads on demand and is browser-cached; recorded voice audio stays on the host device and is not uploaded to a scoring service. If neural inference is unavailable, VoxRelay safely falls back to the local acoustic judge.