A Multilinear Tongue Model Derived from Speech Related MRI Data of the Human Vocal Tract

Alexander Hewer (1, 2, 3) , Stefanie Wuhrer (4) , Ingmar Steiner (1, 3) , Korin Richmond (5)

1 DFKI - Deutsches Forschungszentrum für Künstliche Intelligenz GmbH = German Research Center for Artificial Intelligence
2 Saarbrücken Graduate School of Computer Science
3 MMCI - Cluster of Excellence “Multimodal Computing and Interaction”
4 MORPHEO - Capture and Analysis of Shapes in Motion
5 CSTR - The Centre for Speech Technology Research [Edinburgh]

We present a multilinear statistical model of the human tongue that captures anatomical and tongue pose related shape variations separately. The model is derived from 3D magnetic resonance imaging data of 11 speakers sustaining speech related vocal tract configurations. To extract model parameters, we use a minimally supervised method based on an image segmentation approach and a template fitting technique. Furthermore, we use image denoising to deal with possibly corrupt data, palate surface information reconstruction to handle palatal tongue contacts, and a bootstrap strategy to refine the obtained shapes. Our evaluation shows that, by limiting the degrees of freedom for the anatomical and speech related variations, to 5 and 4, respectively, we obtain a model that can reliably register unknown data while avoiding overfitting effects. Furthermore, we show that it can be used to generate plausible tongue animation by tracking sparse motion capture data.

Mots clés: vocal tract; MRI; tongue; shape analysis; statistical model

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Alexander Hewer, Stefanie Wuhrer, Ingmar Steiner, Korin Richmond. A Multilinear Tongue Model Derived from Speech Related MRI Data of the Human Vocal Tract. Computer Speech and Language, 2018, 51, pp.68-92. ⟨10.1016/j.csl.2018.02.001⟩. ⟨hal-01418460v2⟩