İpek Korkmaz

Masters Student

Bio

İpek was a Master’s student in Data Science and AI at Chalmers University of Technology. Her thesis, “Designing Loss Functions for Learning Sound Timbre Audio Representations in Variational Autoencoders”, explores how variational autoencoders (VAEs) can be trained on synthesis parameters to learn a timbre feature space. As part of this work, she investigated various audio-related loss functions and feature extraction methods to determine the most effective approach for capturing perceptually meaningful timbre representations.