This paper proposes a trajectory model which is based on a mixture density network trained with target features augmented with dynamic features together with an algorithm for estimating maximum likelihood trajectories which respects constraints between the static and derived dynamic features. This model was evaluated on an inversion mapping task. We found the introduction of the trajectory model successfully reduced root mean square error by up to 7.5%, as well as increasing correlation scores.
Cite as: Richmond, K. (2006) A trajectory mixture density network for the acoustic-articulatory inversion mapping. Proc. Interspeech 2006, paper 1790-Mon3WeS.4, doi: 10.21437/Interspeech.2006-213
@inproceedings{richmond06_interspeech,
author={Korin Richmond},
title={{A trajectory mixture density network for the acoustic-articulatory inversion mapping}},
year=2006,
booktitle={Proc. Interspeech 2006},
pages={paper 1790-Mon3WeS.4},
doi={10.21437/Interspeech.2006-213}
}