Talk. See. Share. Instantly.
Voxera is a browser-based real-time communication platform enabling
π voice calls, π₯ video calls, and π₯οΈ screen sharing
using ultra-low latency peer-to-peer streaming.
Inspired by Google Meet, Zoom, and Discord Voice, Voxera is built to explore the core engineering behind real-time systems, without relying on third-party media SDKs.
π Frontend:
π https://voxera-rtc.netlify.app
π Backend (Signaling Server):
π https://voxera-g4st.onrender.com
β οΈ Note: On free hosting, the first request may take ~30 seconds due to cold starts.
β No media server relay
β True peer-to-peer streaming
β Real-time WebSocket signaling
β Clean call lifecycle management
β Production-style UI & UX logic
Voxera is built to understand how real-time systems actually work, not just how to use APIs.
Login β Online Users β Call β Video β Screen Share β End Call
- π One-to-one voice & video calls
- π₯οΈ Screen sharing with system audio
- β‘ Ultra-low latency peer-to-peer streaming
- π Call connect / share / end sounds
- β Synchronized call termination
- π€ Mute / Unmute microphone
- π₯οΈ Start / Stop screen sharing
- π End call
- πͺ Floating camera (Picture-in-Picture)
- β¨ Google Meetβinspired interface
- π Glassmorphism + gradient backgrounds
- π₯ Smart layout (screen-first, camera PiP)
- π Call state overlays (Calling / Incoming)
- π§ Idle onboarding screen
- HTML5
- CSS3 (Glassmorphism, animations)
- JavaScript (ES6+)
- WebRTC (Audio, Video, Screen Capture)
- Socket.IO (Client)
- Node.js
- Express.js
- Socket.IO (WebSockets)
- Google STUN Server
User A ββ WebRTC (Audio / Video / Screen) βββΆ User B
β β²
βββββ Socket.IO (Signaling & Presence) ββββββββ
- User presence & login
- Online user list
- Call requests
- SDP & ICE signaling
- Call end synchronization
- Peer-to-peer media streaming
- Camera & microphone capture
- Screen sharing with system audio
- NAT traversal via ICE
β οΈ Media streams never pass through the server β they flow directly between peers.
Voxera/
β
βββ backend/
β βββ server.js
β βββ package.json
β
βββ frontend/
βββ index.html
βββ style.css
βββ main.js
βββ sounds/
βββ call-connect.mp3
βββ screen-share.mp3
βββ call-end.mp3
git clone https://github.com/vansh070605/Voxera.git
cd Voxeracd backend
npm install
node server.jsβ
Voxera signaling server running on port 3000
- Open
frontend/index.htmlin two browsers/devices - Login with different usernames
- Start calling π
Recommended browser: Google Chrome
Login β Select User β Call β Connect β Share Screen β End Call
β Ending a call on one side disconnects both peers β UI resets cleanly on both ends
- WebRTC peer connection setup
- ICE candidate exchange & NAT traversal
- Media track replacement (screen sharing)
- Real-time state synchronization
- Socket-based signaling architecture
- Production-grade call lifecycle handling
- One-to-one calls only
- No TURN server (may fail on strict NATs)
- No authentication
- No call recording
- No group calls (yet)
- π₯ Group calls / voice rooms
- π₯ Camera on/off toggle
- β±οΈ Call duration analytics
- π Authentication & user profiles
- π TURN server integration
- π Network quality indicators
- π± Improved mobile UX
Voxera is a real-time communication web application built using WebRTC and WebSockets, enabling peer-to-peer voice calls, video calls, and screen sharing with synchronized call lifecycle management and a modern, production-grade UI.
Vansh Agrawal Engineering Student | Full-Stack & Real-Time Systems Enthusiast
Voxera focuses on real-world real-time communication engineering, not just UI demos. It demonstrates the same foundational concepts used in Google Meet, Zoom, and Discord β built completely from scratch.


