WebRTC-Audio-Stream-Example

Node/Socket.io Signaling Server

The signaling server identifies and facilitates negotiations between transceivers before they drop into a peer-to-peer connection. It must be run on a machine that is accessible by both transceivers.

1. npm install
2. Run node ./server.js on a machine accessible by both planned transceivers.
3. Update the IP or URL and port in the index.html call to new Socket(ipaddr, port). These should point to your signaling server.
4. Leave this running.

Electron Transceiver

1. npm install
2. Run npm start.
3. Enter a room name.
4. Select the stream source.
5. Press the Connect button.
6. Audio should start to flow.

Browser Receiver

Serve the required files over file:// or HTTP to use them in the browser. This has been confirmed to work (some receiver only) in Chrome, FireFox, and Chrome Android.

Make sure to npm install or some of the dependencies won't exist.

file:// protocol

1. The simplest test would be loading the file over the file:// protocol.
2. Open index.html in a browser by double-clicking it or dragging it into an open window.

HTTP Server

• Temporarily serve the current directory for testing: python -m SimpleHTTPServer
• Or move the project folder (with node_modules) into a pre-existing webserver setup.

Notes

• Linux and Mac have no support for capturing an audio stream in the browser. They will still function as receivers.
• Beware of the firewall when trying to reach the signaling server. Mine blocked traffic to it. The signaling server is setup to respond with 404 to any HTTP request. You can check if connection is possible in a browser; helpful in verifying connectivity.
• Unmuting the Local audio element can be used to verify that the stream is actually capturing. You should hear an echoey/phasey effect.