diff --git a/setup.py b/setup.py
index b678cee..6253223 100644
--- a/setup.py
+++ b/setup.py
@@ -18,7 +18,7 @@ def gzip_language_data(root, source):
             # If the gz is newer it doesn't need updating
             print("Skipping {}, already compressed".format(jsonfile))
             continue
-        with open(str(jsonfile), 'r') as infileh, gzip.open(str(outfile), 'w') as outfileh:
+        with open(str(jsonfile), 'r', encoding="utf-8") as infileh, gzip.open(str(outfile), 'w') as outfileh:
             outfileh.write(infileh.read().encode("utf-8"))
         print("Compressed {}".format(jsonfile))