Update tts.py

tts.py

@@ -3,6 +3,7 @@ from transformers import SpeechT5ForTextToSpeech, SpeechT5Processor
 import logging
 import numpy as np
 import soundfile as sf
+import torchaudio
 
 # Set up logging
 logging.basicConfig(level=logging.DEBUG)
@@ -31,7 +32,8 @@ def synthesize_speech(text):
 
         # Decode the generated speech and save to an audio file
         waveform = speech.cpu().numpy().flatten()
-        sf.write("output.wav", waveform, 16000)
+        # Use torchaudio to save the waveform
+        torchaudio.save("output.wav", torch.tensor(waveform).unsqueeze(0), 16000)
         logging.info("Audio file saved successfully.")
         return "output.wav"
     except Exception as e: