First commit

app.py

@@ -127,14 +127,17 @@ def extract_faces(video_file):
     ).run(overwrite_output=True)
 
 
-    return output_path
+    return output_path, audio_path
 
 iface = gr.Interface(
     fn=extract_faces,
     inputs=gr.Video(label="Upload or record your video"),
-    outputs=gr.Video(label="Detected Face Only Video"),
+    outputs=[
+        gr.Video(label="Detected Face Only Video"),
+        gr.Audio(label="Extracted Audio (16kHz)", type="filepath"),
+    ],
     title="Face Detector",
-    description="Upload or record a video. We'll crop face regions and return a face-only video."
+    description="Upload or record a video. We'll crop face regions and return a face-only video and its 16kHz audio."
 )
 
 iface.launch()