diff --git a/requirements.txt b/requirements.txt
index 2666c38c..4df335bd 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,3 +1,4 @@
 lxml>=4.1.1
 requests>=2.20.0
+html5lib==1.1
 
diff --git a/safaribooks.py b/safaribooks.py
index 1d23bee3..ebe91957 100755
--- a/safaribooks.py
+++ b/safaribooks.py
@@ -593,9 +593,11 @@ def get_html(self, url):
             )
 
         root = None
+        html_text = response.text
         try:
-            root = html.fromstring(response.text, base_url=SAFARI_BASE_URL)
-
+            if not re.search("<html", response.text, re.I):
+                html_text = etree.tostring(html.html5parser.fromstring(response.text))
+            root = html.fromstring(html_text, base_url=SAFARI_BASE_URL)
         except (html.etree.ParseError, html.etree.ParserError) as parsing_error:
             self.display.error(parsing_error)
             self.display.exit(