13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Deep Architectures for Articulatory Inversion

Benigno Uria (1), Iain Murray (1), Steve Renals (2), Korin Richmond (2)

(1) Institute for Adaptive and Neural Computation, University of Edinburgh, Scotland, UK
(2) Centre for Speech Technology Research, University of Edinburgh, Scotland, UK

We implement two deep architectures for the acoustic-articulatory inversion mapping problem: a deep neural network and a deep trajectory mixture density network. We find that in both cases, deep architectures produce more accurate predictions than shallow architectures and that this is due to the higher expressive capability of a deep model and not a consequence of adding more adjustable parameters. We also find that a deep trajectory mixture density network is able to obtain better inversion accuracies than smoothing the results of a deep neural network. Our best model obtained an average root mean square error of 0.885 mm on the MNGU0 test dataset.

Index Terms: Articulatory inversion, deep neural network, deep belief network, deep regression network, pretraining

Full Paper

Bibliographic reference.  Uria, Benigno / Murray, Iain / Renals, Steve / Richmond, Korin (2012): "Deep architectures for articulatory inversion", In INTERSPEECH-2012, 867-870.