diff --git a/orangecontrib/text/keywords/__init__.py b/orangecontrib/text/keywords/__init__.py
index bdfb44bdd..8dc84f6c4 100644
--- a/orangecontrib/text/keywords/__init__.py
+++ b/orangecontrib/text/keywords/__init__.py
@@ -15,13 +15,15 @@
 from orangecontrib.text import Corpus
 from orangecontrib.text.keywords.mbert import mbert_keywords
 from orangecontrib.text.keywords.rake import Rake
+from orangecontrib.text.language import ISO2LANG
 from orangecontrib.text.preprocess import StopwordsFilter
 
 # all available languages for RAKE
 from orangecontrib.text.vectorization import BowVectorizer
 
-# todo
-RAKE_LANGUAGES = StopwordsFilter.supported_languages()
+# todo: this is a temporary solution since supported_languages now returns
+#  languages as ISO codes - refactor with keywords language refactoring
+RAKE_LANGUAGES = [ISO2LANG[la] for la in StopwordsFilter.supported_languages()]
 # all available languages for YAKE!
 YAKE_LANGUAGE_MAPPING = {
     "Arabic": "ar",