International Journal of Scientific Research and Engineering Development

Parkinson’s is a chronic nervous system condition that progressively affects one’s motor skills and speech abilities. Variations in voice are noticeable at the initial development of Parkinson’s. Therefore, analysis of speech is significant for identifying Parkinson’s at an earlier stage. This paper proposes a machine learning-based approach that employs voice analysis for Parkinson’s detection. Speech processing is involved for deriving relevant features associated with voice. These features include frequency variability, amplitude uncertainty, and noise factors. These features are then employed for training various machine learning models to classify a person as either Parkinson’s patient or healthy individual. These models are then accessed for their performance efficiency with various indicators for accurate prediction. Experimental analysis validates that machine learning models for voice are effective for identifying Parkinson’s patterns. Additionally, the proposed technique is a cost and time efficient diagnostic tool that would help aid doctors for identifying Parkinson’s.

Mr.Sukhuteertha, P.Veena, Sai Srushti P.B, S.Bhanu Begum, "Detection of Parkinson’s Disease Using ML(Voice Recordings)" International Journal of Scientific Research and Engineering Development, V8(6): Page(2763-2768) Nov-Dec 2025. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.