Analysing MRI and ultrasound scans in speech synthesis
Author
View
Keywords
License
Copyright (c) 2020 Acta Medicinae et Sociologica
![Creative Commons License](http://i.creativecommons.org/l/by-nc-nd/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
How To Cite
Abstract
The articulatory speech synthesis is a new trend in producing machine speech which is based on processing visual information related to voice formation. The profound knowledge of static and dynamic geometrical parameters of speech organs plays a fundamental role in the realization of speech synthesis. To visualize these data MRI and ultrasound scans, which have different geometry, could serve as appropriate sources. The pixels of ultrasound frames can conveniently be managed by setting a polar coordinate system, while for the description of MRI frames a Desceartes coordinate system can serve as a start. Since the ultrasound scans, as opposed to MRI, do not show the back part and the apex of the tongue, only partial information is gained on the movement of the tongue. Consequently, it is important and not trivial at all to concert the geometry of MRI and Ultrasound resources. This writing presents a possible way of geometrical transformation.