From aee85fd98f2fb7e2b8446fd5924010707c1efdda Mon Sep 17 00:00:00 2001
From: CyberWon <mail@shuaibo.wang>
Date: Tue, 11 Jun 2024 20:45:35 +0800
Subject: [PATCH] =?UTF-8?q?fix:=20=E4=BF=AE=E5=A4=8D=E5=8F=82=E8=80=83?=
 =?UTF-8?q?=E9=9F=B3=E9=A2=91=E6=B3=84=E9=9C=B2=E9=97=AE=E9=A2=98=20(#1176?=
 =?UTF-8?q?)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

* Update TextPreprocessor.py

* fix: 修复参考音频泄露的问题。

1. 之前变量命名有错误
2. 还存在参考音频泄露问题。

* 调整一下判断纯符号所处的函数，更符号逻辑
---
 GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py b/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py
index 2cc0023..0891227 100644
--- a/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py
+++ b/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py
@@ -56,7 +56,7 @@ class TextPreprocessor:
         
     def preprocess(self, text:str, lang:str, text_split_method:str)->List[Dict]:
         print(i18n("############ 切分文本 ############"))
-        texts = self.replace_consecutive_punctuation(texts)
+        text = self.replace_consecutive_punctuation(text) # 变量命名应该是写错了
         texts = self.pre_seg_text(text, lang, text_split_method)
         result = []
         print(i18n("############ 提取文本Bert特征 ############"))
@@ -94,6 +94,9 @@ class TextPreprocessor:
         for text in _texts:
             # 解决输入目标文本的空行导致报错的问题
             if (len(text.strip()) == 0):
+               continue
+            if not re.sub("\W+", "", text):       
+                # 检测一下，如果是纯符号，就跳过。
                 continue
             if (text[-1] not in splits): text += "。" if lang != "en" else "."