I want to make a real-time audio classification using python. I have a trained deep learning model but I only only give it wav file as an input. I want to make it real-time where it will use microphone as an input. Is that possible? If so, how will I do it?