Protocol for a model-based evaluation of a dynamic acousticto-articulatory inversion method using electromagnetic articulography

Acoustic-to-articulatory maps based on articula-tory models have typically been evaluated in terms of acoustic accuracy, that is, the distance between mapped and observed acoustic parameters. In this paper we present a method that would allow for the evaluation of such maps in the articulatory do-main. The proposed method estimates the parame-ters of Maeda’s articulatory model on the basis of electromagnetic articulograph data, thus producing full midsagittal views of the vocal tract from the po-sitions of a limited number of sensors attached on articulators.