Update app.py

app.py

@@ -158,6 +158,7 @@ def infer(
         en_bert = en_bert.to(device).unsqueeze(0)
         x_tst_lengths = torch.LongTensor([phones.size(0)]).to(device)
         emo = emo.to(device).unsqueeze(0)
+        print(emo)
         del phones
         speakers = torch.LongTensor([hps.data.spk2id[sid]]).to(device)
         audio = (
@@ -203,7 +204,7 @@ if __name__ == "__main__":
     emotional_model = EmotionModel.from_pretrained(emotional_model_name).to(device)
     hps = utils.get_hparams_from_file('Data/BanGDream/configs/config.json')
     net_g = get_net_g(
-        model_path='Data/BanGDream/models/G_63000.pth', version="2.1", device=device, hps=hps
+        model_path='Data/BangDream/models/G_49000.pth', version="2.1", device=device, hps=hps
     )
     speaker_ids = hps.data.spk2id
     speakers = list(speaker_ids.keys())
@@ -222,14 +223,14 @@ if __name__ == "__main__":
                                 with gr.Row():
                                     gr.Markdown(
                                         '<div align="center">'
-                                        f'<img style="width:auto;height:400px;" src="file/image/{name}.png">' 
+                                        f'<img style="width:auto;height:400px;" src="https://mahiruoshi-bangdream-bert-vits2.hf.space/file/image/{name}.png">' 
                                         '</div>'
                                     )
                                 length_scale = gr.Slider(
                                         minimum=0.1, maximum=2, value=1, step=0.01, label="语速调节"
                                     )
                                 emotion = gr.Slider(
-                                    minimum=0, maximum=9, value=0, step=1, label="Emotion"
+                                    minimum=-10, maximum=10, value=0, step=0.1, label="Emotion"
                                 )
                                 with gr.Accordion(label="参数设定", open=False):
                                     sdp_ratio = gr.Slider(