Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

Preliminary Experiments for Automatic Speech Understanding Through Simple Recurrent Networks

M. A. Castano (1), Enrique Vidal (2), F. Casacuberta (2)

(1) Dpto. de Informatica, Universitat Jaume I de Castellon, Spain
(2) Dpto. Sistemas Informaticos y Computation, Universidad Politecnica de Valenciam, Spain

Experiments with Automatic Speech Understanding (ASU) problems were presented in [1], in which the underlying acoustic and syntactic-semantic structure of the task was tried to be automatically learned into a global model. Results indicated that, while a Simple Recurrent Network (SRN) was actually able to capture most of the acoustic features of the ASU task, the syntactic-semantic structure was difficult to learn. We propose here appropriate modifications to this SRN which significantly improve the understanding rates.

Full Paper

Bibliographic reference.  Castano, M. A. / Vidal, Enrique / Casacuberta, F. (1995): "Preliminary experiments for automatic speech understanding through simple recurrent networks", In EUROSPEECH-1995, 1673-1676.