Update app.py

app.py

@@ -4,12 +4,12 @@ import torch
 import torchaudio
 import traceback
 
-def detect_watermark(audio_filepath):
+def detect_watermark(audio_file):
     try:
         # Load the audio file using torchaudio
-        waveform, sample_rate = torchaudio.load(audio_filepath)
+        waveform, sample_rate = torchaudio.load(audio_file.name)  # Use .name attribute for the file path
 
-        # Check waveform dimensions and adjust if necessary
+        # Ensure waveform is 2D (channels, samples). If it's mono, add an axis.
         if waveform.ndim == 2:
             waveform = waveform.unsqueeze(0)  # Add batch dimension
 
@@ -20,19 +20,16 @@ def detect_watermark(audio_filepath):
         # Interpret and return the detection result
         detection_result = "AI-generated" if result else "genuine"
         return f"This audio is likely {detection_result} based on watermark detection."
-    
+
     except Exception as e:
         # Log the error with traceback for debugging
-        error_message = f"An error occurred: {str(e)}"
         error_traceback = traceback.format_exc()
-        full_error_message = f"{error_message}\n\n{error_traceback}"
-        print(full_error_message)  # Consider logging to a file if print statements are not visible
-        return full_error_message
+        return f"Error occurred: {e}\n\n{error_traceback}"
 
-# Define Gradio interface with "filepath" type for audio input
+# Define Gradio interface with "file" type for audio input
 interface = gr.Interface(
     fn=detect_watermark,
-    inputs=gr.Audio(label="Upload your audio", type="filepath"),
+    inputs=gr.Audio(label="Upload your audio", type="file"),
     outputs="text",
     title="Deep Fake Defender: AI Voice Cloning Detection",
     description="Upload an audio file to check if it's AI-generated or genuine."
@@ -40,3 +37,4 @@ interface = gr.Interface(
 
 if __name__ == "__main__":
     interface.launch()
+