import re, json, urllib.request
html=open('/tmp/yq.html',encoding='utf-8').read()
m=re.search(r'"captionTracks":(\[.*?\]),"audioTracks"',html)
if not m:
    raise SystemExit('no captions')
tracks=json.loads(m.group(1))
url=tracks[0]['baseUrl']
url=url.replace('\\u0026','&')
print(url)
data=urllib.request.urlopen(url, timeout=30).read()
open('/tmp/yq.xml','wb').write(data)
print('bytes',len(data))