Update GPT_SoVITS/TTS_infer_pack/TTS.py

GPT_SoVITS/TTS_infer_pack/TTS.py

@@ -287,7 +287,7 @@ class TTS_Config:
             configs: dict = self._load_configs(self.configs_path)
 
         assert isinstance(configs, dict)
-        version = configs.get("version", "v2").lower()
+        version = "v2" #configs.get("version", "v4").lower() # 使用 v4
         assert version in ["v1", "v2", "v3", "v4"]
         self.default_configs[version] = configs.get(version, self.default_configs[version])
         self.configs: dict = configs.get("custom", deepcopy(self.default_configs[version]))
@@ -369,7 +369,7 @@ class TTS_Config:
         return self.config
 
     def update_version(self, version: str) -> None:
-        self.version = version
+        self.version = "v2"#version
         self.languages = self.v1_languages if self.version == "v1" else self.v2_languages
 
     def __str__(self):
@@ -747,8 +747,8 @@ class TTS:
         )
         with torch.no_grad():
             wav16k, sr = librosa.load(ref_wav_path, sr=16000)
-            if wav16k.shape[0] > 160000 or wav16k.shape[0] < 48000:
-                raise OSError(i18n("参考音频在3~10秒范围外，请更换！"))
+            if wav16k.shape[0] > 240000 or wav16k.shape[0] < 48000:
+                raise OSError(i18n("参考音频在3~15秒范围外，请更换！"))
             wav16k = torch.from_numpy(wav16k)
             zero_wav_torch = torch.from_numpy(zero_wav)
             wav16k = wav16k.to(self.configs.device)