Update app.py

app.py

@@ -6,6 +6,7 @@ import scipy.io.wavfile as wav
 import numpy as np
 import wave
 from datasets import load_dataset, Audio, config
+from IPython.display import Audio
 
 # Load the TTS model from the Hugging Face Hub
 checkpoint = "arham061/speecht5_finetuned_voxpopuli_nl"  # Replace with your actual model name
@@ -101,12 +102,7 @@ def text_to_speech(text):
     # Generate audio
     audio_output = generate_audio(text)
 
-    # Save audio as a .wav file
-    from IPython.display import Audio
-
-    audio = Audio(audio_output.numpy(), rate=16000)
-
-    return audio
+    return Audio(audio_output.numpy(), rate=16000)
 
 # Define the Gradio interface
 inputs = gr.inputs.Textbox(label="Enter text in Urdu")