2020/05/29 by Anna Pompili, Rubén Solera-Ureña, Pompili, Anna +13 · 1 citation
Computer Science · Medicine · #Audio and Speech Processing (eess.AS) #FOS: Electrical engineering #Speech Recognition and Synthesis #Speech and Audio Processing #Voice and Speech Disorders #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.14647
openalex publication_date 2020/05/29 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Parkinson's disease (PD) is a progressive degenerative disorder of the\ncentral nervous system characterized by motor and non-motor symptoms. As the\ndisease progresses, patients alternate periods in which motor symptoms are\nmitigated due to medication intake (ON state) and periods with motor\ncomplications (OFF state). The time that patients spend in the OFF condition is\ncurrently the main parameter employed to assess pharmacological interventions\nand to evaluate the efficacy of different active principles. In this work, we\npresent a system that combines automatic speech processing and deep learning\ntechniques to classify the medication state of PD patients by leveraging\npersonal speech-based bio-markers. We devise a speaker-dependent approach and\ninvestigate the relevance of different acoustic-prosodic feature sets. Results\nshow an accuracy of 90.54% in a test task with mixed speech and an accuracy of\n95.27% in a semi-spontaneous speech task. Overall, the experimental assessment\nshows the potentials of this approach towards the development of reliable,\nremote daily monitoring and scheduling of medication intake of PD patients.\n