AudioSuperResolution

Generate higher quality audio from a lower one, using Convolutional neural network.

In this project, we try to improve the quality of audio (from mp3 quality to wav quality) based on the idea of image super resolution. We transform the audio into frequency domain, then stack them as images and train a Convolutional Network to produce the high quality audio input given a lower quality one.
Run: cd into the folder, then exucute "python main.py" to run.