Update app.py

app.py

@@ -88,14 +88,28 @@ def generate_audio(text, audio_ref, temperature, top_P, top_K, audio_seed_input,
     save_path = "output.wav"
 
     # Run the base speaker tts
-    #src_path = "tmp.wav"
-    #chat_tts(text, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input, src_path)
-    src_path = None
-    sample_rate, audio = chat_tts(text, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input, src_path)[0]
+    src_path = "tmp.wav"
+    chat_tts(text, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input, src_path)
     print("Ready for voice cloning!")
-
+    
+    source_se, audio_name = se_extractor.get_se(src_path, tone_color_converter, target_dir='processed', vad=True)
+    print("Get source segment!")
+    
     # Run the tone color converter
     encode_message = "@Hilley"
+    # convert from file
+    tone_color_converter.convert_data(
+        audio_src_path=src_path,
+        src_se=source_se,
+        tgt_se=target_se,
+        output_path=save_path,
+        message=encode_message)
+
+    ''' 
+    # convert from data
+    src_path = None
+    sample_rate, audio = chat_tts(text, temperature, top_P, top_K, audio_seed_input, text_seed_input, refine_text_flag, refine_text_input, src_path)[0]
+    print("Ready for voice cloning!")
     tone_color_converter.convert_data(
         audio=audio,
         sample_rate=sample_rate,
@@ -103,6 +117,7 @@ def generate_audio(text, audio_ref, temperature, top_P, top_K, audio_seed_input,
         tgt_se=target_se,
         output_path=save_path,
         message=encode_message)
+    '''
     print("Finished!")
 
     return "output.wav"