Update app.py

app.py

@@ -60,7 +60,7 @@ def sle(language,text):
             return text
 
 def get_text(text,hps_ms):
-    text_norm = text_to_sequence(text,hps_ms.data.text_cleaners)
+    text_norm = text_to_sequence(text,hps_ms.symbols,hps_ms.data.text_cleaners)
     if hps_ms.data.add_blank:
         text_norm = commons.intersperse(text_norm, 0)
     text_norm = torch.LongTensor(text_norm)
@@ -69,7 +69,7 @@ def get_text(text,hps_ms):
 def create_tts_fn(ort_sess, speaker_id):
     def tts_fn(text , language, n_scale= 0.667,n_scale_w = 0.8, l_scale = 1 ):
         text =sle(language,text)
-        seq = text_to_sequence(text, cleaner_names=hps.data.text_cleaners)
+        seq = text_to_sequence(text,hps.symbols, cleaner_names=hps.data.text_cleaners)
         if hps.data.add_blank:
             seq = commons.intersperse(seq, 0)
         with torch.no_grad():
@@ -152,8 +152,8 @@ if __name__ == '__main__':
                                             input2 = gr.Dropdown(label="Language", choices=lan, value="自动", interactive=True)
                                         with gr.Column():
                                             input1 = gr.TextArea(label="Text", value=example)
-                                            input4 = gr.Slider(minimum=0, maximum=1.0, label="更改噪声比例(noise scale)，以控制情感", value=0.6)
-                                            input5 = gr.Slider(minimum=0, maximum=1.0, label="更改噪声偏差(noise scale w)，以控制音素长短", value=0.668)
+                                            input4 = gr.Slider(minimum=0, maximum=1.0, label="更改噪声比例(noise scale)，以控制情感", value=0.667)
+                                            input5 = gr.Slider(minimum=0, maximum=1.0, label="更改噪声偏差(noise scale w)，以控制音素长短", value=0.8)
                                             input6 = gr.Slider(minimum=0.1, maximum=10, label="duration", value=1)
                                             btnVC = gr.Button("Submit")
                                             output1 = gr.Audio(label="采样率22050")