Evaluating Parkinson's Disease Detection in Anonymized Speech: A Performance and Acoustic Analysis
This work addresses the privacy concerns associated with using speech for Parkinson's disease detection, which is important for patients and healthcare providers.
This paper explores the trade-off between privacy and Parkinson's disease (PD) detection from speech using two anonymizers on Spanish datasets. The kNN-VC anonymizer achieved F1 scores only 3-7% lower than original baselines, demonstrating the viability of privacy-preserving PD detection.
Automatic detection of Parkinson's disease (PD) from speech is a promising non-invasive diagnostic tool, but it raises significant privacy concerns. Speaker anonymization mitigates these risks, but it may suppress the pathological information necessary for PD detection. We assess the trade-off between privacy and PD detection for two anonymizers (STT-TTS and kNN-VC) using two Spanish datasets. STT-TTS provides better privacy but severely degrades PD detection by eradicating prosodic information. kNN-VC preserves macro-prosodic features such as duration and F0 contours, achieving F1 scores only 3-7\% lower than original baselines, demonstrating that privacy-preserving PD detection is viable when using appropriate anonymization. Finally, an acoustic distortion analysis characterizes specific weaknesses in kNN-VC, offering insights for designing anonymizers that better preserve PD information.