Content The dataset include 30 patients with early untreated Parkinson's disease (PD), 50 patients with REM sleep behavior disorder (RBD), which are at high risk developing Parkinson's disease or other synucleinopathies; and 50 healthy controls (HC). All patients were scored clinically by a well-trained professional neurologist with experience in movement disorders. All subjects were examined during a single session with a speech specialist. All subjects performed reading of standardized, phonetically-balanced text of 80 words and monologue about their interests, job, family or current activities for approximately 90 seconds. Speech features were automatically analyzed by Jan Hlavni?ka et al. (2017). The dataset is provided in .xls and .csv format. Acknowledgements Original paper: **Automated analysis of connected speech reveals early biomarkers of Parkinson’s disease in patients with rapid eye movement sleep behaviour disorder** by Jan Hlavni?ka, Roman ?mejla, Tereza Tykalová, Karel ?onka, Ev?en R??i?ka & Jan Rusz @ [nature.com][1] Inspiration Predict a pattern of neurodegeneration in the dataset of speech features obtained from patients with early untreated Parkinson's disease and patients at high risk developing Parkinson's disease. [1]: https://www.nature.com/articles/s41598-017-00047-5