From 88c8bf7b9e545c440dd81e3f161af3d7a04f6bf6 Mon Sep 17 00:00:00 2001
From: passerbya <hanghang3103@163.com>
Date: Tue, 9 Jul 2024 08:22:06 +0800
Subject: [PATCH] =?UTF-8?q?=E6=9B=B4=E6=8D=A2=E5=89=8D=E7=AB=AF=E4=B8=BAWe?=
 =?UTF-8?q?TextProcessing?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 cosyvoice/cli/frontend.py | 23 +++++++----------------
 1 file changed, 7 insertions(+), 16 deletions(-)

diff --git a/cosyvoice/cli/frontend.py b/cosyvoice/cli/frontend.py
index 32e5539..f10f655 100644
--- a/cosyvoice/cli/frontend.py
+++ b/cosyvoice/cli/frontend.py
@@ -21,12 +21,9 @@ import torchaudio.compliance.kaldi as kaldi
 import torchaudio
 import os
 import inflect
-try:
-    import ttsfrd
-    use_ttsfrd = True
-except:
-    print("failed to import ttsfrd, please normalize input text manually")
-    use_ttsfrd = False
+from tn.chinese.normalizer import Normalizer as ZhNormalizer
+from tn.english.normalizer import Normalizer as EnNormalizer
+
 from cosyvoice.utils.frontend_utils import contains_chinese, replace_blank, replace_corner_mark, remove_bracket, spell_out_number, split_paragraph
 
 
@@ -53,14 +50,8 @@ class CosyVoiceFrontEnd:
         self.instruct = instruct
         self.allowed_special = allowed_special
         self.inflect_parser = inflect.engine()
-        self.use_ttsfrd = use_ttsfrd
-        if self.use_ttsfrd:
-            self.frd = ttsfrd.TtsFrontendEngine()
-            ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
-            assert self.frd.initialize('{}/../../pretrained_models/CosyVoice-ttsfrd/resource'.format(ROOT_DIR)) is True, 'failed to initialize ttsfrd resource'
-            self.frd.set_lang_type('pinyin')
-            self.frd.enable_pinyin_mix(True)
-            self.frd.set_breakmodel_index(1)
+        self.zh_tn_model = ZhNormalizer(remove_erhua=False,full_to_half=False)
+        self.en_tn_model = EnNormalizer()
 
     def _extract_text_token(self, text):
         text_token = self.tokenizer.encode(text, allowed_special=self.allowed_special)
@@ -95,8 +86,7 @@ class CosyVoiceFrontEnd:
     def text_normalize(self, text, split=True):
         text = text.strip()
         if contains_chinese(text):
-            if self.use_ttsfrd:
-                text = self.frd.get_frd_extra_info(text, 'input')
+            text = self.zh_tn_model.normalize(text)
             text = text.replace("\n", "")
             text = replace_blank(text)
             text = replace_corner_mark(text)
@@ -107,6 +97,7 @@ class CosyVoiceFrontEnd:
                                                 token_min_n=60, merge_len=20,
                                                 comma_split=False)]
         else:
+            text = self.en_tn_model.normalize(text)
             text = spell_out_number(text, self.inflect_parser)
             texts = [i for i in split_paragraph(text, partial(self.tokenizer.encode, allowed_special=self.allowed_special), "en", token_max_n=80,
                                                 token_min_n=60, merge_len=20,