Update app.py

app.py

@@ -9,8 +9,8 @@ import numpy as np
 
 import ChatTTS
 
-import se_extractor
-from api import ToneColorConverter
+import OpenVoice.se_extractor as se_extractor
+from OpenVoice.api import ToneColorConverter
 import soundfile
 
 print("loading ChatTTS model...")
@@ -80,7 +80,7 @@ tone_color_converter.load_ckpt(f'{ckpt_converter_en}/checkpoint.pth')
 def generate_audio(text, audio_ref, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input):
     save_path = "output.wav"
     
-    if audio_ref is not None:
+    if audio_ref != "" :
       # Run the base speaker tts
       src_path = "tmp.wav"
       text_data = chat_tts(text, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input, src_path)
@@ -108,17 +108,19 @@ def generate_audio(text, audio_ref, temperature, top_P, top_K, audio_seed_input,
 
 
 with gr.Blocks() as demo:
-    gr.Markdown("# Enjoy chatting with your ai friends on website, telegram and so on! (https://linkin.love)")
+    gr.Markdown("# <center>🥳 ChatTTS x OpenVoice 🥳</center>")
+    gr.Markdown("## <center>🌟 Make it sound super natural and switch it up to any voice you want, nailing the mood and tone also!🌟 </center>")
+    gr.Markdown("### <center>💕 Slide on over to [linkin.love](https://linkin.love) for a chat-fest with your AI buds!💕</center>")
 
     default_text = "Today a man knocked on my door and asked for a small donation toward the local swimming pool. I gave him a glass of water."        
     text_input = gr.Textbox(label="Input Text", lines=4, placeholder="Please Input Text...", value=default_text)
 
 
     default_refine_text = "[oral_2][laugh_0][break_6]"    
-    refine_text_checkbox = gr.Checkbox(label="Refine text：'oral' means add filler words, 'laugh' means add laughter, and 'break' means add a pause. (0-10) ", value=True)
+    refine_text_checkbox = gr.Checkbox(label="Refine text", info="'oral' means add filler words, 'laugh' means add laughter, and 'break' means add a pause. (0-10) ", value=True)
     refine_text_input = gr.Textbox(label="Refine Prompt", lines=1, placeholder="Please Refine Prompt...", value=default_refine_text)
     with gr.Column():    
-        voice_ref = gr.Audio(label="请上传您喜欢的语音文件", type="filepath", value="")
+        voice_ref = gr.Audio(label="Reference Audio", info="Click on the ✎ button to upload your own target speaker audio", type="filepath", value="examples/speaker.mp3")
 
     with gr.Row():
         temperature_slider = gr.Slider(minimum=0.00001, maximum=1.0, step=0.00001, value=0.3, label="Audio temperature")
@@ -148,15 +150,12 @@ with gr.Blocks() as demo:
                           inputs=[text_input, voice_ref, temperature_slider, top_p_slider, top_k_slider, audio_seed_input, text_seed_input, refine_text_checkbox, refine_text_input], 
                           outputs=[audio_output,text_output])
 
-parser = argparse.ArgumentParser(description='ChatTTS demo Launch')
+parser = argparse.ArgumentParser(description='ChatTTS-OpenVoice Launch')
 parser.add_argument('--server_name', type=str, default='0.0.0.0', help='Server name')
 parser.add_argument('--server_port', type=int, default=8080, help='Server port')
 args = parser.parse_args()
 
-    # demo.launch(server_name=args.server_name, server_port=args.server_port, inbrowser=True)
-
-
-
+# demo.launch(server_name=args.server_name, server_port=args.server_port, inbrowser=True)
 
 if __name__ == '__main__':
     demo.launch()