Workshop 2002 - SuperSID: Exploiting High-level Information for High-performance Speaker Recognition

SuperSID: Exploiting High-Level Information for High-Performance Speaker Recognition
Project Description Final Presentation Slide Show (PDF slides) Fusion update Final Report

	Name	Affiliation
	Reynolds, Doug *	MIT Lincoln Laboratory
	Peskin, Barbara	ICSI
	Navratil, Jiri	IBM
	Campbell, Joe	MIT Lincoln Laboratory
	Andrews, Walter	DoD
	Klusacek, David	Charles University
	Adami, Andre	OGI
	Jin, Qin	Carnegie Mellon University
	Abramson, Joy	York University
	Mihaescu, Radu	Princeton University
	Goddfrey, Jack**	DoD
	Jones, Doug**	MIT Lincoln Laboratory
	Xiang, Bing**	BBN Tecnhologies
	* Group Leader
	**Off site Participants

Opening Day Progress Report, 31 July 2002 Progress Report, 14 August 2002 David Klusacek's “Pronunciation Modeling” Proposal Andre Adami's “Prosodic: Modeling Pitch/Energy Dynamics” Proposal Qin Jin's “High-Level Phonetic Information for Speaker Recognition” Proposal
Reading List SuperSID Tutorial, Doug Reynolds SuperSID Listening Quiz Lab Exercise, Jiri Navratil Idiolect Lab, Walt Andrews & Joe Campbell Integrating High-Level Information, Larry Heck The Manifold Advantages of Articulatory Rep., John Hogden Prosodic Features for the Switchboard Database: SRI Database, Luciana Ferrer
The SuperSID Project: Exploiting High-level Information for High-accuracy Speaker Recognition Phonetic Speaker Recognition Using Maximum-Likelihood Binary-Decision Tree Models Conditional Pronunciation Modeling in Speaker Detection Using Prosodic and Conversational Features for High-Performance Speaker Recognition: Report from JHU'02 Modeling Prosodic Dynamics for Speaker Recognition Combining Cross-streams and Time Dimensions in Phonetic Speaker Recognition
Fusing High- and Low-Level Features for Speaker Recognition Segmentation of Speech for Speaker and Language Recognition

The Center for Language and Speech Processing The Johns Hopkins University 3400 North Charles Street, Barton Hall Baltimore, MD 21218
	Telephone: (410) 516-4237		Fax: (410) 516-5050		E-mail: clsp@clsp.jhu.edu

CLSP Homepage : Workshop Homepage

Research Group	Friday, October 24, 2003