This is a small page accompanying a bachelor thesis on kick drum generation, full text. Here you can listen to some samples from each of the models. We tried to pick samples that were representative of the quality that can be expected when running these models on random input data, but the quality ofcourse varies from sample to sample.

Dataset

Augmented dataset

We describe our data augmentation method in section 3.2 of the report.

WaveGAN

Our baseline model is the WaveGAN model, trained on the original dataset:


We then trained the same model, but on the augmented dataset to see if 'more' data would improve its output:


In section 4.1.3 we introduce our progressively growing versions of WaveGAN. The first two below are the $\alpha$-fading version, and the second two are the $\eta$-fading version.

Nistal PGAN

We compared our Progressive WaveGAN to a more standard PGAN architecture. The first two below are the 'mag-if' representation, and the next two are the 'complex' representation.

WaveRNN and WaveNet

Now to some failed experiments, the two samples below are from WaveRNN (left) and WaveNet (right).

SpecGAN