Update app.py

app.py

@@ -2,6 +2,8 @@ import streamlit as st
 import librosa
 import numpy as np
 import onnxruntime as ort
+import os
+import requests
 
 # Audio padding function
 def pad(x, max_len=64600):
@@ -24,13 +26,29 @@ def preprocess_audio_segment(segment, cut=64600):
     segment = pad(segment, max_len=cut)
     return np.expand_dims(np.array(segment, dtype=np.float32), axis=0)  # Add batch dimension
 
+# Download ONNX model from Hugging Face
+def download_model(url, local_path="RawNet_model.onnx"):
+    """
+    Download the ONNX model from a URL if it doesn't already exist locally.
+    """
+    if not os.path.exists(local_path):
+        with st.spinner("Downloading ONNX model..."):
+            response = requests.get(url)
+            if response.status_code == 200:
+                with open(local_path, "wb") as f:
+                    f.write(response.content)
+                st.success("Model downloaded successfully!")
+            else:
+                raise Exception("Failed to download ONNX model")
+    return local_path
+
 # Sliding window prediction function
-def predict_with_sliding_window(audio_path, onnx_model_url, window_size=64600, step_size=64600, sample_rate=16000):
+def predict_with_sliding_window(audio_path, onnx_model_path, window_size=64600, step_size=64600, sample_rate=16000):
     """
     Use a sliding window to predict if the audio is real or fake over the entire audio.
     """
     # Load ONNX runtime session
-    ort_session = ort.InferenceSession(onnx_model_url)
+    ort_session = ort.InferenceSession(onnx_model_path)
 
     # Load audio file
     waveform, _ = librosa.load(audio_path, sr=sample_rate)
@@ -69,22 +87,25 @@ st.write("Upload an audio file to detect if it is Real or Fake.")
 # File uploader
 uploaded_file = st.file_uploader("Upload your audio file (WAV or MP3)", type=["wav", "mp3"])
 
-if uploaded_file is not None:
-    # Path to your ONNX model
-    onnx_model_url = "https://huggingface.co/Mrkomiljon/DeepVoiceGuard/resolve/main/RawNet_model.onnx"
+# ONNX model URL (replace with your actual Hugging Face model URL)
+onnx_model_url = "https://huggingface.co/Mrkomiljon/DeepVoiceGuard/resolve/main/RawNet_model.onnx"
 
+# Ensure ONNX model is downloaded locally
+onnx_model_path = download_model(onnx_model_url)
+
+if uploaded_file is not None:
     # Save uploaded file temporarily
     with open("temp_audio_file.wav", "wb") as f:
         f.write(uploaded_file.read())
 
     # Perform prediction
     with st.spinner("Processing..."):
-        result, avg_probability = predict_with_sliding_window("temp_audio_file.wav", onnx_model_url)
+        result, avg_probability = predict_with_sliding_window("temp_audio_file.wav", onnx_model_path)
 
     # Display results
     st.success(f"Prediction: {result}")
     st.info(f"Confidence: {avg_probability:.2f}%")
 
     # Clean up temporary file
-    import os
     os.remove("temp_audio_file.wav")
+