Axel’s current research interest focuses on advancing deep learning techniques for tasks related to voice and music processing, analysis, and transformation. This includes neural vocoding, the exploration of signal representation and manipulation within latent spaces, as well as the investigation of disentangling strategies in these latent spaces. The research results are regularly used for artistic or commercial media productions.
Expertise keywords
Digital Signal Processing | Signal Models | Neural Vocoding | Sound Representation and Transformation with Deep Neural Networks