Update app.py

app.py

@@ -3,6 +3,7 @@ from transformers import AutoFeatureExtractor, AutoModelForAudioClassification
 import numpy as np
 import torch
 from torch.nn.functional import softmax
+import soundfile as sf
 
 # Path to the local directory where the model files are stored within the Space
 local_model_path = "./"
@@ -23,9 +24,12 @@ def predict_voice(audio_file):
     """
     # Gradio passes the audio file as a tuple (file_name, file_path). We only need the file_path.
     audio_file_path = audio_file[1]
+    
+    # Load the audio file. Adjust the loading mechanism based on your audio file format.
+    waveform, sample_rate = sf.read(audio_file_path)
 
     # Convert the input audio file to model's expected format.
-    inputs = extractor(audio_file_path, return_tensors="pt", sampling_rate=extractor.sampling_rate)
+    inputs = extractor(waveform, return_tensors="pt", sampling_rate=sample_rate)
     
     # Generate predictions from the model.
     with torch.no_grad():  # Ensure no gradients are calculated
@@ -45,6 +49,7 @@ def predict_voice(audio_file):
     result = f"The voice is classified as '{label}' with a confidence of {confidence:.2f}%."
     return result
 
+
 # Setting up the Gradio interface
 iface = gr.Interface(
     fn=predict_voice,