tumbach/titleturtle

Special tool for parsing metadata from audio streams

titleturtle

What is it?

Titleturtle -- special tool for parsing metadata from audio streams.

How it works?

Titleturtle reads tags like artist and title directly from an audio stream, which contains Icecast metadata.

Install

npm install   # or `yarn install`
cp config.js.example config.js

# change сonfig.js

npm start     # or `yarn start` or `node app`

Usage

let websocket = new WebSocket("ws://localhost:8080/ws");
    websocket.onopen = () => {
      websocket.send('HISTORY tumbach 5'); // request 5 last tracks <optional>
      websocket.send('SUB tumbach'); // subscribe to tag updates
    };
    websocket.onmessage = (e) => {
      try {
        let data = JSON.parse(e.data);
        console.log(data);
      } catch (e) {
        //
      }
    };
    window.onunload = () => websocket.close();

Known issues

• Titleturtle is experiencing some issues during reconnections (see issue #1).
• Titleturtle can not choose a stream by the lowest bitrate for now. This feature would save bandwidth.

titleturtle's Issues

Track info stops being updated

Track info stops updating under some conditions:

• titleturtle stops receiving the stream
• server stops streaming
• HTML pages are returned by Icecast/nginx (404/502/etc.)

It is necessary to implement listening to these events and restarting metadata parsing.

Follow redirects

At this moment, titleturtle does not support 3xx HTTP codes.
Example:
https://radio.tumba.ch/stream-ultra-low => https://radio.tumba.ch/stream-ultra-low.aac