Fixing english tokenization

2023-10-06 17:32:21 +02:00 · 2023-10-06 17:32:21 +02:00 · 1ec341857e
parent 3063846aa7
commit 1ec341857e
1 changed files with 3 additions and 2 deletions
--- a/TTS/tts/layers/xtts/tokenizer.py
+++ b/TTS/tts/layers/xtts/tokenizer.py
@ -223,8 +223,9 @@ class VoiceBpeTokenizer:
            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":