Multimodal Fusion of Electromagnetic, Ultrasound and MRI Data for Building an Articulatory Model
Michaël Aron, Marie‐Odile Berger, Erwan Kerrien, Inria Nancy Grand-Est · 2008
Data fusion from multiple sensors is of significant interest to the speech research community, as it can potentially provide a better picture of speech produc-tion through the use of complementary sensor modal-ities. This paper deals with the practical aspects of this problem, such as acquisition and processing of the dynamic ultrasound (US) and electromagnetic (EM) data of the tongue during speech production, static MRI images of the vocal tract using repeti-tions, and registration of the data from these differ-ent sources to a common reference frame. To the best of our knowledge, this is the first work that demon-strates the potential of static and dynamic data fusion in the construction of articulatory databases. 1