1 code implementation • 14 Jul 2019 • Viktor Tóth, Lauri Parkkonen
To overcome these obstacles, we developed a novel class of SS methods, by training deep recurrent autoencoders for image-to-sound conversion.