In this paper, we argue the possibility of retrieving the vocal tract dynamics from speech as was produced by a given speaker. First, we present an inversion method that provides a complete set of articulatory solutions without excessive constraints. The selection of articulatory trajectories is not an easy task, as many trajectories are possible. Based on some inversion experiments and data observation, we show the importance of the introduction of the notion of speaker articulatory strategy to be able to model this variability.
Cite as: Ouni, S. (2005) Can we retrieve vocal tract dynamics that produced speech? toward a speaker articulatory strategy model. Proc. Interspeech 2005, 1037-1040, doi: 10.21437/Interspeech.2005-419
@inproceedings{ouni05_interspeech,
author={Slim Ouni},
title={{Can we retrieve vocal tract dynamics that produced speech? toward a speaker articulatory strategy model}},
year=2005,
booktitle={Proc. Interspeech 2005},
pages={1037--1040},
doi={10.21437/Interspeech.2005-419}
}