This chapter describes current methods and future trends in signal processing and pattern recognition techniques applied to the automatic analysis of motor impairments observed in patients with Parkinson’s disease (PD). A revision of the state-of-the-art is presented considering methods and applications that include the analysis of speech, gait, and handwriting signals for the evaluation of PD. The suitability of existing methods and the potential use of novel approaches are also presented in the context of the automatic diagnosis and monitoring of the disease progression. Apart from the classical clinical scales used to assess the neurological state of PD patients, specific scales designed to evaluate the speech impairments of the patients are also reviewed. Additionally, the suitability of a modified version of the Frenchay Dysarthria Assessment scale (m-FDA) is evaluated. Besides the classical signal processing methods based on the analysis in the time and frequency domains, other methods that consider non-linear features and also deep neural networks are explored in several experiments..