Update app.py

app.py

@@ -1,4 +1,4 @@
-import gradio as gr
+mport gradio as gr
 import librosa
 import numpy as np
 import tensorflow as tf
@@ -10,14 +10,14 @@ model = load_model('voice_authentication_model.keras')  # Replace with your mode
 # Function to extract MFCC features and make a prediction
 def predict_user_or_non_user(audio):
     try:
-        # Load the audio file using librosa, force resampling to 16000 Hz
-        y, sr = librosa.load(audio, sr=16000)  # Ensuring correct sampling rate (16000 Hz)
+        # Load the audio file using librosa
+        y, sr = librosa.load(audio, sr=16000)  # sr=None to keep the original sampling rate
         
         # Optional: Normalize the audio volume (if necessary)
         y = librosa.util.normalize(y)
         
         # Extract MFCC features from the audio
-        mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)
+        mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)  # You can adjust n_mfcc as needed
         mfccs = np.mean(mfccs.T, axis=0)  # Take the mean of MFCCs over time to reduce dimension
         
         # Reshape the MFCCs to match the input shape expected by the model
@@ -30,6 +30,7 @@ def predict_user_or_non_user(audio):
         print(f"Raw Prediction Output: {prediction}")
         
         # Apply thresholding based on the raw prediction value
+        # If the model outputs a probability, try adjusting the threshold (e.g., 0.6 instead of 0.5)
         if prediction[0] > 0.5:
             return "User"
         else:
@@ -41,10 +42,10 @@ def predict_user_or_non_user(audio):
 # Define the Gradio interface
 iface = gr.Interface(
     fn=predict_user_or_non_user,  # The function to call when an audio input is given
-    inputs=gr.Audio(type="numpy"),  # Using numpy to handle live audio input
+    inputs=gr.Audio(type="filepath"),  # Corrected audio input setup
     outputs="text",  # Output will be text (User or Non-user)
     live=True  # Live mode so the interface updates in real-time
 )
 
 # Launch the interface
-iface.launch()
+iface.launch()