Audio Super Resolution in Python3 with Tensorflow 1.5.0 (ref. https://kuleshov.github.io/audio-super-res/)