Conditional Generative Data Augmentation for Clinical Audio Datasets

no code implementations • 22 Mar 2022 • Matthias Seibold, Armando Hoch, Mazda Farshad, Nassir Navab, Philipp Fürnstahl

In this work, we propose a novel data augmentation method for clinical audio datasets based on a conditional Wasserstein Generative Adversarial Network with Gradient Penalty (cWGAN-GP), operating on log-mel spectrograms.

Data Augmentation Generative Adversarial Network