Automatic acoustic speech analysis is a promising tool to reliably capture speech changes as it can be used to distinguish neurotypical from pathological speech. Furthermore, it provides non-invasive monitoring of treatment response at low-burden and low-cost. Our aim is to monitor levodopa-related speech changes in Parkinson’s disease and to establish a digital speech-based biomarker that detects these changes.

The project aims to expand the current knowledge of speech motor deficits in people with Parkinson's disease in order to develop speech-related biomarkers for diagnostic and therapeutic speech interventions. The aim is to determine the earliest time of speech motor changes in isolated REM sleep behavior disorder and Parkinson's disease in order to enable early detection of both diseases by means of automated speech recognition and to be able to initiate early therapy.

This project is funded with 100.000 € by the German Parkinson Foundation.

Learn more about the project in this informational video.

This project is part of the third funding period of the CRC 1252 “Prominence in Language” at the University of Cologne.

The project aims to transfer foundational knowledge about phonetic-prosodic changes in the speech system of neurotypical and pathological speakers into a clinical context. Together with our industry partner ki:elements GmbH, we are developing an algorithm that reliably detects clinically relevant speech changes in Parkinson’s disease using digital biomarkers. The algorithm is intended for long-term use in clinical practice to help preserve the communication abilities of Parkinson's patients through timely speech therapy. Further information can be found here.