ISCA Archive Interspeech 2013
ISCA Archive Interspeech 2013

Training an articulatory synthesizer with continuous acoustic data

Santitham Prom-on, Peter Birkholz, Yi Xu

This paper reports preliminary results of our effort to address the acoustic-to-articulatory inversion problem. We tested an approach that simulates speech production acquisition as a distal learning task, with acoustic signals of natural utterances in the form of MFCC as input, VocalTractLab . a 3D articulatory synthesizer controlled by target approximation models as the learner, and stochastic gradient descent as the training method. The approach was tested on a number of natural utterances, and the results were highly encouraging.


doi: 10.21437/Interspeech.2013-98

Cite as: Prom-on, S., Birkholz, P., Xu, Y. (2013) Training an articulatory synthesizer with continuous acoustic data. Proc. Interspeech 2013, 349-353, doi: 10.21437/Interspeech.2013-98

@inproceedings{promon13_interspeech,
  author={Santitham Prom-on and Peter Birkholz and Yi Xu},
  title={{Training an articulatory synthesizer with continuous acoustic data}},
  year=2013,
  booktitle={Proc. Interspeech 2013},
  pages={349--353},
  doi={10.21437/Interspeech.2013-98}
}