Dysarthric Speech Classification Using Hierarchical Multilayer Perceptrons and Posterior Rhythmic Features Sid-Ahmed Selouani, Habiba Dahmani, Riadh Amami, and Habib Hamam Abstract. In this paper class posterior distributions are combined with a hierarchal structure of multilayer Perceptrons to perform an automatic assessment of dysarthric speech. In addition to the standard Mel-frequency coefficients, this hybrid classifier uses rhythm-based features as input parameters since the preliminary evidence from perceptual experiments show that rhythm troubles may be the common character- istic of various types of dysarthria. The Nemours database of American dysarthric speakers is used throughout experiments. Results show the relevance of rhythm met- rics and the effectiveness of the proposed hybrid classifier to discriminate the levels of dysarthria severity. 1 Introduction Dysarthria is linked to the disturbance of brain and nerve stimuli of the muscles in- volved in the production of speech. This impairment induces disturbances in timing and accuracy of movements necessary for prosodically normal, efficient and intelli- gible speech. Rhythm troubles may be the common characteristic of various types of dysarthria, but all types of dysarthria affect the articulation of consonants and vow- els (in very severe dysarthria) leading to a slurring speech [4]. Even if the rhythm is identified as the main feature that characterizes dysarthria, assessment methods are mainly based on perceptual evaluations. Despite their numerous advantages that Sid-Ahmed Selouani · Habib Hamam Universit´ e de Moncton, New Brunswick, Canada e-mail: selouani@umcs.ca,habib.hamam@umoncton.ca Habiba Dahmani INRS-Universit´ e du Qu´ ebec, Montreal, Canada e-mail: dahmani@emt.inrs.ca Riadh Amami ´ Ecole ESPRIT, Tunis, Tunisia e-mail: riadhamami@gmail.com E. Corchado et al. (Eds.): SOCO 2011, AISC 87, pp. 437–444, 2011. springerlink.com c  Springer-Verlag Berlin Heidelberg 2011