Background: Parkinson's is a neurological disorder not as rare as projected by maestros in this field, most commonly as it is considered a taboo in general societal norms or rebuffed as an old age syndrome. The complexity in detection resulting from the lack of reliable tests at the early stages of PD makes it hard to deal with chronic stages. Methods: This paper focuses on using extracted features from speech signals to detect an abnormality and deducing whether the patient is suffering from PD. Glottal Closure Instants (GCI) detection algorithms such as Speech Event Detection using Residual Excitation and the Mean Based Signal (SEDREAMS) is compared with electroglottographic (EGG) recordings as reference ground. The speech samples are taken from the UCI Machine Learning repository, which contains multiple speakers' recordings. Results: The early onset of PD can be predicted at benign stages of the disorder and help in treatment at later stages of Parkinson's disease. SEDREAMS showed an accuracy of 88.7%. Conclusion: SEDREAMS is used to diagnose Parkinson’s disease early with precise detection of Glottal closing and opening instants.