Simple js script to fetch the auto-generated youtube captions for the video currently open in your browser window

fetch-yt-captions.js

function decodeUnicodeEscapeSequence(str) {
  return str
    .replace(/\\u/g, "%u")
    .replace(/(%u)([a-fA-F\d]{4})/gi, function(_, _, hex) {
      return String.fromCharCode(parseInt(hex, 16));
    });
}

function parseXML(xmlStr) {
  // Create a new DOMParser object
  var parser = new DOMParser();

  // Parse the XML string into a DOM Document
  var doc = parser.parseFromString(xmlStr, "application/xml");

  // Find all <text> elements in the document
  var texts = Array.from(doc.querySelectorAll("text"));

  // Extract the text content of each <text> element and store it in an array
  var textContents = texts.map((textElement) => textElement.textContent.trim());

  return textContents;
}

// Fetch url for auto-generated captions from youtube
async function getCaptionUrl() {
  const response = await fetch(window.location);
  rawHtml = await response.text();

  let matches = rawHtml.match(
    new RegExp('(?<=captionTracks.*baseUrl":")[^"]+"', "g"),
  );

  if (!matches) {
    return;
  }
  var decodedUrl = decodeUnicodeEscapeSequence(matches[0].replace('"', ""));
  return decodedUrl;
}


async function fetchCaptions() {
  const url = await getCaptionUrl();
  console.log(url);
  const response = await fetch(url);
  rawXml = await response.text();
  var captions = parseXML(rawXml);
  console.log(`No of Captions: ${captions.length}`);
  return captions;
}

var captions = await fetchCaptions();

Warning

Do not paste code into your dev console if you don't trust the author/don't understand the code. This can be a security risk.

How to use

1. Copy the above code
2. Go to the tab which has a YouTube video open
3. Open the dev console (F12)
4. Paste the code you copied in step 1 and hit enter
5. The variable captions contains a list of all captions for the YouTube video currently open in your browser window