diff --git a/newspaper/outputformatters.py b/newspaper/outputformatters.py
index 56bea437..dc9d7610 100644
--- a/newspaper/outputformatters.py
+++ b/newspaper/outputformatters.py
@@ -8,10 +8,14 @@
 __copyright__ = 'Copyright 2014, Lucas Ou-Yang'
 
 from HTMLParser import HTMLParser
+import logging
 
 from .text import innerTrim
 
 
+log = logging.getLogger(__name__)
+
+
 class OutputFormatter(object):
 
     def __init__(self, config):
@@ -60,7 +64,12 @@ def get_formatted(self, top_node):
     def convert_to_text(self):
         txts = []
         for node in list(self.get_top_node()):
-            txt = self.parser.getText(node)
+            try:
+                txt = self.parser.getText(node)
+            except ValueError:  # lxml error
+                log.warning('Error parsing lxml node', exc_info=True)
+                txt = None
+
             if txt:
                 txt = HTMLParser().unescape(txt)
                 txt_lis = innerTrim(txt).split(r'\n')