Update app.py

app.py

@@ -58,7 +58,7 @@ def clone_and_speak(text, speaker_wav):
 
     # Generate base TTS voice
         speed = 1.0
-        source_se = torch.load(f'checkpoints/base_speakers/EN/{speaker_key}.pth', map_location=device)
+        #source_se = torch.load(f'checkpoints/base_speakers/EN/{speaker_key}.pth', map_location=device)
         model.tts_to_file(text, speaker_id, tmp_melo_path,speed=speed)
 
     # Use speaker_wav as reference to extract style embedding
@@ -68,7 +68,7 @@ def clone_and_speak(text, speaker_wav):
     # Run the tone conversion
         tone_color_converter.convert(
         audio_src_path=tmp_melo_path,
-        src_se=source_se,
+        src_se=ref_se,
         tgt_se=ref_se,
         output_path=final_output_path,
         message="@HuggingFace",