Update app.py

app.py

@@ -50,24 +50,24 @@ def clone_and_speak(text, speaker_wav):
     # Use English speaker model
     model = TTS(language="EN", device=device)
     speaker_ids = model.hps.data.spk2id
-    #default_speaker_id = next(iter(speaker_ids.values()))
+    default_speaker_id = next(iter(speaker_ids.values()))
 
     for speaker_key in speaker_ids.keys():
         speaker_id = speaker_ids[speaker_key]
         speaker_key = speaker_key.lower().replace('_', '-')
 
     # Generate base TTS voice
-        speed = 1.0
+    speed = 1.0
         #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)
+    model.tts_to_file(text, speaker_id, tmp_melo_path,speed=speed)
 
     # Use speaker_wav as reference to extract style embedding
-        from openvoice import se_extractor
-        torch.hub.load('snakers4/silero-vad', 'silero_vad', force_reload=False)
-        ref_se, _ = se_extractor.get_se(speaker_wav, tone_color_converter, vad=True)
+    from openvoice import se_extractor
+    torch.hub.load('snakers4/silero-vad', 'silero_vad', force_reload=False)
+    ref_se, _ = se_extractor.get_se(speaker_wav, tone_color_converter, vad=True)
 
     # Run the tone conversion
-        tone_color_converter.convert(
+    tone_color_converter.convert(
         audio_src_path=tmp_melo_path,
         src_se=ref_se,
         tgt_se=ref_se,