Main features
- Video, audio and screen sharing with virtual backgrounds
- Simulcast and Dynacast with H264, VP8, VP9 and AV1
- Collaborative whiteboard with PDF, DOCX and PPTX upload
- Shared notepad, polls and breakout rooms
- Optional end-to-end encryption for media and data
- MP4 recording and RTMP/RTMPS broadcasting
- RTMP and WHIP ingress for streaming software
- Telephone dial-in through a SIP gateway
- Live transcription, spoken translation and summaries via your own AI provider
- API with PHP and JavaScript SDKs
- WordPress, Moodle and Joomla plugins, plus LTI
- Custom branding and interface changes without editing core code
What plugNmeet does
plugNmeet is a web conferencing system from MynaParrot SL, designed to be plugged into other websites and applications rather than used only as a standalone meeting site. It is built from three parts: a Go server (the repository tracked here), a React client and a separate recorder, all on top of the LiveKit selective forwarding unit. LiveKit supplies simulcast, Dynacast and modern codecs including AV1.
The meeting tools lean toward teaching and presentations. Next to video, audio, screen sharing and virtual backgrounds there is a collaborative whiteboard that accepts PDF, Word and PowerPoint files, a shared notepad, polls and breakout rooms. Meetings can be recorded to MP4 or broadcast over RTMP or RTMPS, and streams from OBS or similar software can be brought in through RTMP or WHIP. A SIP gateway adds telephone dial-in.
Integrations and AI
Integration is the main reason to pick plugNmeet. Its API comes with PHP and JavaScript SDKs, there are ready-made plugins for WordPress, Moodle and Joomla, and LTI covers other learning platforms. Branding and the interface can be customized without changing core code, and mobile apps can be built around it through its hybrid integration.
The AI agent offers live transcription, spoken translation and meeting summaries. It does not run models itself: the operator configures accounts with Azure, Google Gemini or OpenAI, so audio and text go to that provider when the features are on.
Security
Media is encrypted in transit with WebRTC by default. The README describes optional end-to-end encryption for both media and data channels, which means the media server forwards content it cannot read.
Self-hosting plugNmeet
The documentation recommends the installation script on a clean Ubuntu or Debian LTS server with a public IP address and no other web server running. You need two subdomains, one for the app and one for TURN, and an email address for Let’s Encrypt. For production the documented minimum is 4 CPU cores and 4 GB of RAM, rising to 8 or more cores and 8 GB or more when the recorder runs on the same machine, with around 100 Mbit/s of bandwidth; 2 cores and 2 to 4 GB are enough for testing.
Docker images and a sample Compose file are also provided. The sample shows the full set of services: the API server, Redis, MariaDB, NATS, the LiveKit server, LiveKit ingress and LiveKit SIP. Manual installs need LiveKit, Redis and MariaDB or MySQL, plus LibreOffice and mupdf-tools to convert office files for the whiteboard.
Hosted version
MynaParrot runs plugNmeet Cloud, a paid managed service, and a public demo is available at demo.plugnmeet.com.
Limitations
plugNmeet has a smaller community than Jitsi or BigBlueButton and depends on one company. A full installation involves many moving parts, from LiveKit to NATS and the recorder. The AI features rely on cloud AI providers, and there are no ready-made native apps.
Who it suits
plugNmeet suits developers, LMS administrators and site owners who want to add meetings with classroom tools to WordPress, Moodle, Joomla or their own application under the MIT license. For a classroom tool with a longer history in education, compare BigBlueButton; for an embeddable platform with LiveKit-compatible SDKs, see OpenVidu.
License details
MIT. The server is MIT; the client and recorder live in separate repositories. AI features use external providers configured by the operator.
Technical details
Self-hosting needs
- LiveKit
- Redis
- MariaDB or MySQL
- NATS
Built with
- Go
- React
- TypeScript
- LiveKit
GitHub topics
- ai
- ai-agent
- communication
- livekit
- pion
- video-chat
- video-chat-meetings
- video-conferencing
- web-conferencing
- webrtc
- webrtc-call
- webrtc-meetings
- webrtc-video
Reviewed by the OpenSourcePicks editors from the project's repository and documentation. Something out of date? Tell us.