Classification of Parkinson Disease Based on Analysis and Synthesis of Voice Signal

The most important application of voice profiling is pathological detection. Parkinson's disease a chronic neurological degenerative affecting the central nervous system responsible for essentially progressive evolution movement disorders. 70% to 90% Parkinson’s (PD) patients show an affected voice. This paper proposes methodology PD based on acoustic, glottal, physical, and electrical parameters. results that acoustic parameter more in case as compared glottal physical authors achieved 97.2% accuracy differentiate Parkinson healthy using jitter pitch ratio proposed algorithm. Authors also algorithm poles calculation vocal tract find formants tract. Further, are used finding transfer function filter. In end, suggested parameters model changed voices.