diff --git a/tests/metrics/de/test_reference_free_text_quality.py b/tests/metrics/de/test_reference_free_text_quality.py
index 1c754ddc..bce1dbff 100644
--- a/tests/metrics/de/test_reference_free_text_quality.py
+++ b/tests/metrics/de/test_reference_free_text_quality.py
@@ -30,7 +30,7 @@ def test_sentiment_openai(generated_outputs):
     mock_chat_completion = Mock(spec=ChatCompletion)
     mock_chat_completion.choices = [
         Mock(message=Mock(function_call=Mock(
-            arguments="{\n  \"sentiment\": \"Positive\"\n}")))
+            arguments="{\n  \"sentiment\": \"Positiv\"\n}")))
     ]
 
     # Calling the openai.resources.chat.Completions.create method requires an
diff --git a/tests/metrics/de/test_source_based_text_quality.py b/tests/metrics/de/test_source_based_text_quality.py
index d4e0a7c0..561af4dd 100644
--- a/tests/metrics/de/test_source_based_text_quality.py
+++ b/tests/metrics/de/test_source_based_text_quality.py
@@ -39,7 +39,7 @@ def test_factual_consistency_openai(generated_outputs, sources):
     mock_chat_completion = Mock(spec=ChatCompletion)
     mock_chat_completion.choices = [
         Mock(message=Mock(function_call=Mock(
-            arguments="{\n  \"factuality\": \"Fully Consistent\"\n}")))
+            arguments="{\n  \"factuality\": \"Vollständig Konsistent\"\n}")))
     ]
 
     # Calling the openai.resources.chat.Completions.create method requires an