Update xtts_train.py

xtts_train.py

@@ -44,6 +44,12 @@ def run_tts(lang, tts_text, speaker_audio_file):
     if XTTS_MODEL is None or not speaker_audio_file:
         return "You need to run the previous step to load the model !!", None, None
 
+    # Access the list element
+    speaker_audio_file = speaker_audio_file[0]
+
+    # Remove brackets and quotes
+    speaker_audio_file = speaker_audio_file.replace("[", "").replace("]", "").replace("'", "")
+    
     gpt_cond_latent, speaker_embedding = XTTS_MODEL.get_conditioning_latents(audio_path=speaker_audio_file, gpt_cond_len=XTTS_MODEL.config.gpt_cond_len, max_ref_length=XTTS_MODEL.config.max_ref_len, sound_norm_refs=XTTS_MODEL.config.sound_norm_refs)
     out = XTTS_MODEL.inference(
         text=tts_text,