[bug] fix some bugs

app.py

@@ -12,6 +12,7 @@ from torch import no_grad, LongTensor
 import webbrowser
 import logging
 import gradio.processing_utils as gr_processing_utils
+from gradio_client import utils as client_utils
 logging.getLogger('numba').setLevel(logging.WARNING)
 limitation = os.getenv("SYSTEM") == "spaces"  # limit text and audio length in huggingface spaces
 
@@ -20,7 +21,7 @@ def audio_postprocess(self, y):
     data = audio_postprocess_ori(self, y)
     if data is None:
         return None
-    return gr_processing_utils.encode_url_or_file_to_base64(data["name"])
+    return client_utils.encode_url_or_file_to_base64(data["name"])
 gr.Audio.postprocess = audio_postprocess
 
 def get_text(text, hps):
@@ -35,7 +36,7 @@ def vits(text, language, speaker_id, noise_scale, noise_scale_w, length_scale):
     if not len(text):
         return "输入文本不能为空！", None, None
     text = text.replace('\n', ' ').replace('\r', '').replace(" ", "")
-    if len(text) > 100 and limitation:
+    if len(text) > 200 and limitation:
         return f"输入文字过长！{len(text)}>100", None, None
     if language == 0:
         text = f"[ZH]{text}[ZH]"
@@ -92,7 +93,7 @@ download_audio_js = """
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
     parser.add_argument('--device', type=str, default='cpu')
-    parser.add_argument('--api', action="store_true", default=False)
+    parser.add_argument('--api', action="store_true", default=True)
     parser.add_argument("--share", action="store_true", default=False, help="share gradio app")
     parser.add_argument("--colab", action="store_true", default=False, help="share gradio app")
     args = parser.parse_args()
@@ -107,6 +108,7 @@ if __name__ == '__main__':
         **hps_ms.model)
     _ = net_g_ms.eval().to(device)
     speakers = hps_ms.speakers
+    speakers = [f"{i}.{s}" for i, s in enumerate(speakers)]
     model, optimizer, learning_rate, epochs = utils.load_checkpoint(r'./model/G_953000.pth', net_g_ms, None)
     
     with gr.Blocks() as app:
@@ -118,16 +120,16 @@ if __name__ == '__main__':
             with gr.TabItem("vits"):
                 with gr.Row():
                     with gr.Column():
-                        input_text = gr.Textbox(label="Text (100 words limitation) " if limitation else "Text", lines=5, value="今天晚上吃啥好呢。", elem_id=f"input-text")
+                        input_text = gr.Textbox(label="Text (200 words limitation) " if limitation else "Text", lines=5, value="可莉不知道喔。", elem_id=f"input-text")
                         lang = gr.Dropdown(label="Language", choices=["中文", "日语", "中日混合（中文用[ZH][ZH]包裹起来，日文用[JA][JA]包裹起来）"],
                                     type="index", value="中文")
                         btn = gr.Button(value="Submit")
                         with gr.Row():
                             search = gr.Textbox(label="Search Speaker", lines=1)
                             btn2 = gr.Button(value="Search")
-                        sid = gr.Dropdown(label="Speaker", choices=speakers, type="index", value=speakers[228])
+                        sid = gr.Dropdown(label="Speaker", choices=speakers, type="index", value=speakers[329])
                         with gr.Row():
-                            ns = gr.Slider(label="noise_scale(控制感情变化程度)", minimum=0.1, maximum=1.0, step=0.1, value=0.6, interactive=True)
+                            ns = gr.Slider(label="noise_scale(控制感情变化程度)", minimum=0.1, maximum=1.0, step=0.1, value=0.1, interactive=True)
                             nsw = gr.Slider(label="noise_scale_w(控制音素发音长度)", minimum=0.1, maximum=1.0, step=0.1, value=0.668, interactive=True)
                             ls = gr.Slider(label="length_scale(控制整体语速)", minimum=0.1, maximum=2.0, step=0.1, value=1.2, interactive=True)
                     with gr.Column():