Research

Human communication arises from complex interactions between cognition, motor control, and the physical processes of speech production and perception. Our research aims to better understand these processes and to develop computational methods that support clinical assessment and rehabilitation.

Our work focuses on several overarching goals:

• developing quantitative models of speech and voice production
• identifying acoustic and physiological markers of communication disorders
• applying machine learning methods to speech and voice analysis
• translating computational approaches into clinically meaningful tools

Through interdisciplinary collaboration between engineering, medicine, and speech science, we aim to contribute to a deeper understanding of communication disorders and their treatment.

Selected research projects

Voice conversion for the processing of pathological speech

This ongoing research project investigates voice conversion techniques for the processing of pathological speech signals. The aim is to develop computational approaches that can improve the analysis and intelligibility of speech produced by individuals with communication disorders.

More information: FWF Research Radar

Objective differentiation of dysphonic voice quality types

This completed clinical research project investigated objective methods for differentiating perceptual voice quality types in dysphonic patients. The project aimed to connect clinical perceptual evaluation with quantitative acoustic analysis in order to improve the reliability of voice diagnostics.

More information: FWF Research Radar