International audienceWe introduce a method for predicting midsagittal contours of orofacial organs during speech and swallowing from dynamic MRI images. For each speaker, a set of 60 representative images for which contours have been manually traced allows for training PCA images. The data serve to derive articulatory contour PCA models and a multilinear model that predicts contour parameters from image parameters. The obtained contours are then corrected by a modified Active Shape Model (ASM) using the local information of the pixel intensity profiles along the contour normals. The performance of this method (mean “points to contour” errors between 0.57 and 0.70 mm) is insensitive to the type of MRI sequence (conventional gradient echoes ...
Quantification of the anatomic and functional aspects of the tongue is pertinent to analyse the mech...
The mechanisms of speech production are complex and have been raising attention from researchers of ...
International audienceWe describe a minimally-supervised method for computing a statistical shape sp...
International audienceWe introduce a method for predicting midsagittal contours of orofacial organs ...
International audienceWe introduce a method for predicting midsagittal contours of orofacial articul...
We introduce a method for predicting midsagittal contours of orofacial articulators from real-time M...
Speech production mechanisms can be characterized at a peripheral level by both their acoustic and a...
International audienceThis work concerns the construction of articulatory models for synthesis of sp...
International audienceMRI allows to characterize the shape and position of speech articulators, but ...
Abstract Volume images of tongue, jaw, velum, nasopharyngeal wall, etc., were acquired by MRI on one...
International audienceArticulatory speech synthesis requires generating realistic vocal tract shapes...
International audienceArticulatory speaker normalisation based on MRI-data using three-way linear de...
DIn this paper 1, we propose a shape-based variational framework to curve evolution for the segmenta...
Quantification of the anatomic and functional aspects of the tongue is pertinent to analyse the mech...
The mechanisms of speech production are complex and have been raising attention from researchers of ...
International audienceWe describe a minimally-supervised method for computing a statistical shape sp...
International audienceWe introduce a method for predicting midsagittal contours of orofacial organs ...
International audienceWe introduce a method for predicting midsagittal contours of orofacial articul...
We introduce a method for predicting midsagittal contours of orofacial articulators from real-time M...
Speech production mechanisms can be characterized at a peripheral level by both their acoustic and a...
International audienceThis work concerns the construction of articulatory models for synthesis of sp...
International audienceMRI allows to characterize the shape and position of speech articulators, but ...
Abstract Volume images of tongue, jaw, velum, nasopharyngeal wall, etc., were acquired by MRI on one...
International audienceArticulatory speech synthesis requires generating realistic vocal tract shapes...
International audienceArticulatory speaker normalisation based on MRI-data using three-way linear de...
DIn this paper 1, we propose a shape-based variational framework to curve evolution for the segmenta...
Quantification of the anatomic and functional aspects of the tongue is pertinent to analyse the mech...
The mechanisms of speech production are complex and have been raising attention from researchers of ...
International audienceWe describe a minimally-supervised method for computing a statistical shape sp...