Parkinson's disease (PD) is one of the most common neurodegenerative disorders. PD is referred as idiopathic, that is, as having no known cause; its main symptoms are tremor, rigidity and general loss of muscle control. Research shows that speech may be a useful indicator for discriminating patients with PD from healthy controls. The paper describes our contribution to the INTERSPEECH 2015 Special Session “Computational Paralinguistics Challenge (ComParE): Parkinson's Condition Sub-Challenge”. The main goal of the challenge is to perform automatic classification (regression) on speech produced by patients with Parkinson's disease. The paper presents our method of linear regression models on a set of extracted acoustic features from the middle of vowels in words, sentences and continuous speech, and the partitioning of the speech samples according to their total length into parts with long, medium and short duration.