Assessing privacy-friendly local open-source voice annotation for participants with Parkinson’s disease

Conference item

There is significant potential clinical benefit to be gained in capturing symptom data from individuals with Parkinson’s Disease (PD). For this purpose, sensor data is often collected. However, labels (ground truth) data is also beneficial, both to train (supervised learning) and to validate outcomes from automated monitoring systems. With the increasing use of voice assistants, this modality has been proposed for labelling. In this study, we examine some design patterns for voice-agent-supported labelling, identify failure modes, and make use of the MDVR-KCL dataset to benchmark a widely used key component, a speech-to-text pipeline. We identify that this component shows rapid increase in several error metrics (WER, CER, WIL) when employed on data from mildly symptomatic participants. We identify some potential mitigating steps and discuss potential future work.

Authors: Tonkin, EL; Tourte, GJL

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer
• Host title: Annotation of Real-World Data for Artificial Intelligence Systems: 9th International Workshop, ARDUOUS 2025, Bologna, Italy, October 25–26, 2025, Proceedings
• Pages: 159–176
• Series: Communications in Computer and Information Science
• Series number: 2706
• Place of publication: Cham, Switzerland
• Publication date: 2025-10-24
• Acceptance date: 2025-07-11
• Event title: 9th International Workshop on Annotation of Real World Data for Artificial Intelligent Systems (ARDUOUS 2025)
• Event location: Bologna, Italy
• Event website: https://arduous.eu/
• Event start date: 2025-10-25
• Event end date: 2025-10-26
• DOI: 10.1007/978-3-032-09117-8_10
• EISSN: 1865-0937
• ISSN: 1865-0929
• EISBN: 9783032091178
• ISBN: 9783032091161
• Language: English
• Keywords: Parkinson’s disease; privacy-first engineering; voice assistants