Audiovisual-to-articulatory speech inversion using HMMs

Katsamanis, A; Papandreou, G; Maragos, P

dc.contributor.author	Katsamanis, A	en
dc.contributor.author	Papandreou, G	en
dc.contributor.author	Maragos, P	en
dc.date.accessioned	2014-03-01T02:44:29Z
dc.date.available	2014-03-01T02:44:29Z
dc.date.issued	2007	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/31843
dc.subject	Canonical Correlation Analysis	en
dc.subject	Linear Model	en
dc.subject	Point of View	en
dc.subject	Posterior Probability	en
dc.subject	Speech Production	en
dc.subject	Audio Video	en
dc.subject	Hidden Markov Model	en
dc.subject	Point of Interest	en
dc.subject.other	Hidden Markov models	en
dc.subject.other	Markov processes	en
dc.subject.other	Signal processing	en
dc.subject.other	Technical presentations	en
dc.subject.other	Audio visuals	en
dc.subject.other	Audiovisual speeches	en
dc.subject.other	Canonical Correlation Analysis	en
dc.subject.other	Electromagnetic articulography	en
dc.subject.other	Points of interests	en
dc.subject.other	Speech productions	en
dc.subject.other	Statistical frameworks	en
dc.subject.other	Visual modalities	en
dc.subject.other	Visual speeches	en
dc.subject.other	Vocal tracts	en
dc.subject.other	Speech recognition	en
dc.title	Audiovisual-to-articulatory speech inversion using HMMs	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/MMSP.2007.4412915	en
heal.identifier.secondary	http://dx.doi.org/10.1109/MMSP.2007.4412915	en
heal.identifier.secondary	4412915	en
heal.publicationDate	2007	en
heal.abstract	We address the problem of audiovisual speech inversion, namely recovering the vocal tract's geometry from auditory and visual speech cues. We approach the problem in a statistical framework, combining ideas from multistream Hidden Markov Models and canonical correlation analysis, and demonstrate effective estimation of the trajectories followed by certain points of interest in the speech production system. Our experiments show that exploiting both audio and visual modalities clearly improves performance relative to either audio-only or visual-only estimation. We report experiments on the QSMT database which contains audio, video, and electromagnetic articulography data recorded in parallel. © 2007 IEEE.	en
heal.journalName	2007 IEEE 9Th International Workshop on Multimedia Signal Processing, MMSP 2007 - Proceedings	en
dc.identifier.doi	10.1109/MMSP.2007.4412915	en
dc.identifier.spage	457	en
dc.identifier.epage	460	en