diff --git a/youtube_dl/extractor/radiobremen.py b/youtube_dl/extractor/radiobremen.py
index 6d130d3d9..9f7e6af15 100644
--- a/youtube_dl/extractor/radiobremen.py
+++ b/youtube_dl/extractor/radiobremen.py
@@ -5,10 +5,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_duration
 
 
 class RadioBremenIE(InfoExtractor):
-    _VALID_URL = r'http?://(?:www\.)?radiobremen\.de/mediathek/(index\.html)?\?id=(?P<video_id>[0-9]+)'
+    _VALID_URL = r'http?://(?:www\.)?radiobremen\.de/mediathek/(?:index\.html)?\?id=(?P<id>[0-9]+)'
     IE_NAME = 'radiobremen'
 
     _TEST = {
@@ -16,6 +17,7 @@ class RadioBremenIE(InfoExtractor):
         'info_dict': {
             'id': '114720',
             'ext': 'mp4',
+            'duration': 1685,
             'width': 512,
             'title': 'buten un binnen vom 22. Dezember',
             'description': 'Unter anderem mit diesen Themen: 45 Flüchtlinge sind in Worpswede angekommen +++ Freies Internet für alle: Bremer arbeiten an einem flächendeckenden W-Lan-Netzwerk +++ Aktivisten kämpfen für das Unibad +++ So war das Wetter 2014 +++',
@@ -23,32 +25,32 @@ class RadioBremenIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
 
         meta_url = "http://www.radiobremen.de/apps/php/mediathek/metadaten.php?id=%s" % video_id
         meta_doc = self._download_webpage(meta_url, video_id, 'Downloading metadata')
         title = self._html_search_regex("<h1.*>(?P<title>.+)</h1>", meta_doc, "title")
         description = self._html_search_regex("<p>(?P<description>.*)</p>", meta_doc, "description")
-        duration = self._html_search_regex("L&auml;nge:</td>\s+<td>(?P<duration>[0-9]+:[0-9]+)</td>", meta_doc, "duration")
+        duration = parse_duration(
+            self._html_search_regex("L&auml;nge:</td>\s+<td>(?P<duration>[0-9]+:[0-9]+)</td>", meta_doc, "duration"))
 
         page_doc = self._download_webpage(url, video_id, 'Downloading video information')
         pattern = "ardformatplayerclassic\(\'playerbereich\',\'(?P<width>[0-9]+)\',\'.*\',\'(?P<video_id>[0-9]+)\',\'(?P<secret>[0-9]+)\',\'(?P<thumbnail>.+)\',\'\'\)"
         mobj = re.search(pattern, page_doc)
-        width, video_id, secret, thumbnail = int(mobj.group("width")), mobj.group("video_id"), mobj.group("secret"), mobj.group("thumbnail")
-        video_url = "http://dl-ondemand.radiobremen.de/mediabase/{:}/{:}_{:}_{:}.mp4".format(video_id, video_id, secret, width)
+        video_url = (
+            "http://dl-ondemand.radiobremen.de/mediabase/%s/%s_%s_%s.mp4" %
+            (video_id, video_id, mobj.group("secret"), mobj.group('width')))
 
+        formats = [{
+            'url': video_url,
+            'ext': 'mp4',
+            'width': int(mobj.group("width")),
+        }]
         return {
             'id': video_id,
             'title': title,
             'description': description,
             'duration': duration,
-            'formats': [
-                {'url': video_url,
-                 'ext': 'mp4',
-                 'width': width,
-                 'protocol': 'http'
-                 }
-            ],
-            'thumbnail': thumbnail,
+            'formats': formats,
+            'thumbnail': mobj.group('thumbnail'),
         }