Show simple item record

dc.contributor.authorFrankel, Joe
dc.contributor.authorKing, Simon
dc.date.accessioned2006-05-09T11:44:59Z
dc.date.available2006-05-09T11:44:59Z
dc.date.issued2005
dc.identifier.citationIn Proceedings, Interspeech'2005 - Eurospeech, 9th European Conference on Speech Communication and Technology, Lisbon, Portugal, September 4-8, 2005en
dc.identifier.urihttp://www.isca-speech.org/archive/interspeech_2005
dc.identifier.urihttp://hdl.handle.net/1842/926
dc.description.abstractArtificial neural networks (ANN) have proven to be well suited to the task of articulatory feature (AF) recognition. Previous studies have taken a cascaded approach where separate ANNs are trained for each feature group, making the assumption that features are statistically independent. We address this by using ANNs to provide virtual evidence to a dynamic Bayesian network (DBN). This gives a hybrid ANN/DBN model and allows modelling of inter-feature dependencies. We demonstrate significant increases in AF recognition accuracy from modelling dependencies between features, and present the results of embedded training experiments in which a set of asynchronous feature changes are learned. Furthermore, we report on the application of a Viterbi training scheme in which we alternate between realigning the AF training labels and retraining the ANNs.en
dc.format.extent147299 bytes
dc.format.extent78714 bytes
dc.format.mimetypeapplication/postscript
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherInternational Speech Communication Associationen
dc.subjectArtificial neural networksen
dc.subjectspeech recognitionen
dc.subjectarticulatory feature recognitionen
dc.subjectdynamic Bayesian networken
dc.titleA hybrid ANN/DBN approach to articulatory feature recognitionen
dc.typeConference Paperen


Files in this item

This item appears in the following Collection(s)

Show simple item record