[F] Fix #251: Expected a value of type 'str'

system/tts_engines/f5tts/model_engine.py

@@ -756,7 +756,7 @@ class tts_class:
     ):
         """Process text and prepare for batch inference"""
         # Split the input text into batches
-        audio, sr = torchaudio.load(ref_audio)
+        audio, sr = torchaudio.load(str(ref_audio))
         max_chars = int(len(ref_text.encode("utf-8")) / (audio.shape[-1] / sr) * (25 - audio.shape[-1] / sr))
         gen_text_batches = self.chunk_text(gen_text, max_chars=max_chars)
         

system/tts_engines/xtts/model_engine.py

@@ -1113,7 +1113,7 @@ class tts_class:
                 else:
                     self.print_message("Starting non-streaming generation", message_type="debug_tts")
                     output = self.model.inference(**common_args)
-                    torchaudio.save(output_file, torch.tensor(output["wav"]).unsqueeze(0), 24000)
+                    torchaudio.save(str(output_file), torch.tensor(output["wav"]).unsqueeze(0), 24000)
                     self.print_message(f"Saved audio to: {output_file}", message_type="debug_tts")
 
             elif self.current_model_loaded.startswith("apitts"):