Fixing english tokenization

coqui-ai · Oct 6, 2023 · 1ec3418 · 1ec3418
1 parent 3063846
commit 1ec3418
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/TTS/tts/layers/xtts/tokenizer.py b/TTS/tts/layers/xtts/tokenizer.py
@@ -223,8 +223,9 @@ def preprocess_text(self, txt, lang):
             results = kks.convert(txt)
             txt = " ".join([result["kana"] for result in results])
             txt = basic_cleaners(txt)
-        elif lang == "en":
-            txt = english_cleaners(txt)
+        # elif lang == "en":
+        #     txt = english_cleaners(txt)
+        # English cleaner remove the language tag [en]
         elif lang == "ar":
             txt = arabic_cleaners(txt)
         elif lang == "zh-cn":