This is an investigation carried out for the evaluation of the classification performance (Speech, Music, and Noise) of 1D (WaveNet) and 2D (MobileNet) CNN and RNN (GRU) on the MUSAN corpus.
Link to Corpus: http://www.openslr.org/17/
For results and more info see "report.pdf"