From 75c7f1d971f08f09a2a83f20b118ac2436e465c9 Mon Sep 17 00:00:00 2001 From: CyberWon Date: Mon, 10 Jun 2024 17:34:55 +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=E7=9A=84=E9=97=AE=E9=A2=98?= =?UTF-8?q?=E3=80=82?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 1. 之前变量命名有错误 2. 还存在参考音频泄露问题。 --- 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 12bf2d8b..826b41a5 100644 --- a/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py +++ b/GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py @@ -56,11 +56,14 @@ class TextPreprocessor: def preprocess(self, text:str, lang:str, text_split_method:str)->List[Dict]: print(i18n("############ 切分文本 ############")) - text = self.replace_consecutive_punctuation(text) + 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 ############")) for text in tqdm(texts): + if not re.sub("\W+", "", text): + # 检测一下，如果是纯符号，就跳过。 + continue phones, bert_features, norm_text = self.segment_and_extract_feature_for_text(text, lang) if phones is None: continue