From 8c0cb0d691554d8311d8904972ae3efa8bfd1cc4 Mon Sep 17 00:00:00 2001
From: Ella Zhang <144317607+EllaZhangCA@users.noreply.github.com>
Date: Tue, 23 Sep 2025 02:50:39 -0700
Subject: [PATCH] =?UTF-8?q?=E4=B8=BA=E4=B8=AD=E6=96=87=E6=B7=BB=E5=8A=A0?=
 =?UTF-8?q?=E4=BA=86=E8=B4=A7=E5=B8=81=E8=AE=A1=E6=95=B0?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 GPT_SoVITS/text/cleaner.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/GPT_SoVITS/text/cleaner.py b/GPT_SoVITS/text/cleaner.py
index 7c13023c..388b4e16 100644
--- a/GPT_SoVITS/text/cleaner.py
+++ b/GPT_SoVITS/text/cleaner.py
@@ -13,7 +13,7 @@ from text import symbols2 as symbols_v2
 
 special = [
     # ("%", "zh", "SP"),
-    ("￥", "zh", "SP2"),
+    # ("￥", "zh", "SP2"), #加了货币计数所以人民币符不是SP2了
     ("^", "zh", "SP3"),
     # ('@', 'zh', "SP4")#不搞鬼畜了，和第二版保持一致吧
 ]
@@ -31,7 +31,7 @@ def clean_text(text, language, version=None):
     if language not in language_module_map:
         language = "en"
         text = " "
-    if language in ("zh"): #处理货币似乎只能这里截胡，不然货币符号会被吞
+    if language in ("zh"): #处理货币似乎最佳方案是这里截胡，不然可能被吞...
         from text.zh_normalization.num import (
             RE_CNY_PREFIX, RE_CNY_SUFFIX, replace_cny_prefix, replace_cny_suffix,
             RE_USD_SYMBOL, RE_USD_SUFFIX, replace_usd_symbol, replace_usd_suffix,)