diff --git a/GPT_SoVITS/inference_webui.py b/GPT_SoVITS/inference_webui.py
index 3924ff65..b925a382 100644
--- a/GPT_SoVITS/inference_webui.py
+++ b/GPT_SoVITS/inference_webui.py
@@ -214,7 +214,7 @@ def resample(audio_tensor, sr0):
 ###todo:put them to process_ckpt and modify my_save func (save sovits weights), gpt save weights use my_save in process_ckpt
 #symbol_version-model_version-if_lora_v3
 from process_ckpt import get_sovits_version_from_path_fast,load_sovits_new
-def change_sovits_weights(sovits_path,prompt_language=None,text_language=None):
+def change_sovits_weights(sovits_path,prompt_language=None,text_language=None, cat=None):
     print("the sovits model is updated!")
     global vq_model, hps, version, model_version, dict_language,if_lora_v3
     version, model_version, if_lora_v3=get_sovits_version_from_path_fast(sovits_path)
@@ -254,7 +254,7 @@ def change_sovits_weights(sovits_path,prompt_language=None,text_language=None):
     else:
         hps.model.version = "v2"
     version=hps.model.version
-    # print("sovits版本:",hps.model.version)
+    print("sovits版本:",hps.model.version)
     if model_version!="v3":
         vq_model = SynthesizerTrn(
             hps.data.filter_length // 2 + 1,
@@ -279,6 +279,7 @@ def change_sovits_weights(sovits_path,prompt_language=None,text_language=None):
     else:
         vq_model = vq_model.to(device)
     vq_model.eval()
+
     if if_lora_v3==False:
         print("loading sovits_%s"%model_version,vq_model.load_state_dict(dict_s2["weight"], strict=False))
     else: