Normalize punctuation before German TTS

2026-08-24 19:26:07 +02:00
parent faba681153
commit dec4709ae9
2 changed files with 23 additions and 2 deletions
@@ -110,6 +110,21 @@ class LanguageSegmentationTests(unittest.TestCase):
                self.assertIn("29 Kilometer pro Stunde", spoken)
                self.assertNotIn("km", spoken.lower())

+    def test_visual_punctuation_becomes_natural_pauses(self):
+        spoken = gateway.clean_for_speech(
+            "Status: stabil – keine Fehler; Docker-Container laufen."
+        )
+        self.assertEqual(
+            spoken,
+            "Status, stabil, keine Fehler, Docker Container laufen.",
+        )
+        self.assertNotRegex(spoken, r"[:;\-‐‑‒–—−]")
+
+    def test_punctuation_only_segments_are_never_synthesized(self):
+        segments = gateway.prepare_segments("Status: – alles läuft.")
+        self.assertTrue(segments)
+        self.assertTrue(all(any(char.isalnum() for char in part) for _, part in segments))
+
    def test_weather_summary_is_split_into_short_complete_chunks(self):
        text = (
            "Heute in Rastatt: teils sonnig, trocken, 10 bis 22 Grad. "