A deep learning framework for Speech-Music discrimination of continuous audio streams - View it on GitHub