Curriculum Vitae: Thomas K. Harris
Research Associate III
School of Computer Science
Carnegie Mellon University
Pittsburgh, PA 15213
E-mail: Thomas.Harris@cs.cmu.edu
Research
- Multi-Agent Dialogue
- What issues arise when multiple heterogeneous spoken dialogue
agents interact with a human interlocutor? I'm developing
multiple-agent spoken dialogue systems to run on teams of heterogeneous
robots.
- Speech Graffiti
- What universal interaction primitives are necessary (or at least
broadly useful) of any spoken language dialogue system? And how can we
characterize and exploit the nature of spoken interactions between
people and machines? I'm developing systems and testing theories
related to these questions.
- Grammatical Language Modeling for Speech Recognition and Spoken
Language Understanding
- How can grammatical knowlege improve speech recognition and spoken
language understanding? I'm investigating some effects of statistical
repesentations of grammars with a modified version of Sphinx-3.4.
Education
- M.S. in Language Technologies, Carnegie Mellon University, 2004.
Thesis: The Speech Graffiti Personal Universal Controller: A Speech Interface for Appliances
- B.S. in Computer Science, George Mason University, 2001.
Employment
- 2006-present: Reseach Associate III, Carnegie Mellon University, Pittsburgh, PA.
- 2001-2005: Graduate Research Assistant, Carnegie Mellon University, Pittsburgh, PA.
- 1999-2000: Senior Software Engineer, Saraf Software Inc., Falls Church, VA.
- 1998: Software Engineer, TSI, Chantilly, VA.
- 1995-1997: Co-Founder and IT Manager, Internet Doorway, Inc., Jackson, MS.
Publications
Journal Papers
Conference Papers
- Thomas K. Harris and Alexander I. Rudnicky. TeamTalk: A
platform for multi-human-robot dialog research in coherent real and
virtual spaces. (2007) Association for the Advancement of
Artificial Intelligence, Vancouver B.C., Canada. PDF
- Dan Bohus, Antoine Raux, Thomas K. Harris, Maxine Eskanazi, and
Alexander I. Rudnicky. Olympus: an open-source framework for
conversational spoken language interface research. (2007)
Bridging the Gap: Academic and Industrial Research in Dialog
Technology, Rochester, New York. PDF
- M. Bernardine Dias, Thomas K. Harris, Brett Browning, E. Gil
Jones, Brenna Argall, Manuela Veloso, Anthony Stenz, Alex
Rudnicky. Dynamically Formed Human-Robot Teams Performing
Coordinated Tasks. (2006) AAAI Spring Symposium: To Boldly Go
Where No Human-Robot Team Has Gone, Palo Alto, California. PDF
- Thomas K. Harris, Satanjeev Banerjee, and Alexander
I. Rudnicky. Heterogeneous Multi-Robot Dialogues for Search
Tasks. (2005) AAAI Spring Symposium: Dialogical Robots,
Palo Alto, California. PDF
- Thomas K. Harris and Roni Rosenfeld. A Universal Speech
Interface for Appliances. (2004) Proc. International
Conference on Speech and Language Processing, Jeju, Korea. PS.GZ
- Thomas K. Harris, Satanjeev Banerjee, Alexander Rudnicky, June
Sison, Kerry Bodine, and Alan Black. A Research Platform for
Multi-Agent Dialogue Dynamics. (2004) Proceedings of The IEEE
International Workshop on Robotics and Human Interactive
Communications, Kurashiki, Japan. PDF
- Jeffrey Nichols, Brad A. Myers, Kevin Litwack, Michael Higgins,
Joseph Hughes, and Thomas K. Harris. Describing Appliance User
Interfaces Abstractly with XML. (2004) Proc. Workshop on
Developing User Interfaces with XML: Advances on User Interface
Description Languages, Gallipoli, Italy. PDF
- Brad A. Myers, Jeffrey Nichols, Jacob O. Wobbrock, Kevin Litwack,
Michael Higgins, Joe Hughes, Thomas K. Harris, Roni Rosenfeld, and
Mathilde Pignol. Handheld Devices for Control. (2003).
Proc. Human-Computer Interaction Consortium, Winter Park,
Colorado. PDF
- Jeffrey Nichols, Brad A. Myers, Michael Higgins, Joseph Hughes,
Thomas K. Harris, Roni Rosenfeld, and Kevin Litwack. Personal
Universal Controllers: Controlling Complex Appliances with GUIs and
Speech. (2003). Extended Abstracts of Computer-Human
Interaction, Ft. Lauderdale, Florida. PDF
- Jeffrey Nichols, Brad A. Myers, Thomas K. Harris, Roni Rosenfeld,
Stefanie Shriver, Michael Higgins, and Joseph Hughes. Requirements
for Automatically Generating Multi-Modal Interfaces for Complex
Appliances. (2002). Proc. IEEE International Conference on
Multimodal Interfaces, Pittsburgh, Pennsylvania. PDF
- Arthur Toth, Thomas K. Harris, James Sanders, Stefanie Shriver,
and Roni Rosenfeld. Towards Every-Citizen's Speech Interface: An
Application Generator for Speech Interfaces to Databases.
(2002). Proc. International Conference on Spoken Language
Processing, Denver, Colorado. PS.GZ
- Jeffrey Nichols, Brad A. Myers, Michael Higgins, Joseph Hughes,
Thomas K. Harris, Roni Rosenfeld, and Mathilde Pignol. Generating
Remote Control Interfaces for Complex Appliances. (2002). Proc.
Symposium on User Interface Software & Technology, Paris,
France. PDF
Presentations
- TeamTalk: Human-Robot Group Communication. September 17, 2006. Demo Session, Interspeech 2006. Pittsburgh, Pennsylvania.
- Heterogenous Multi-Robot Dialogues for Search Tasks. March
22, 2005. AAAI
Spring Symposium: Dialogical Robots. Palo Alto, California. PPT
- Multi-Agent Dialogue. June 8, 2004. Boeing Phantom Works
multi-robot kick-off meeting. Pittsburgh, Pennsylvania. PPT
- Conversational Game Theory. November 25, 2003. Graduate Seminar
on Dialog Processing. Pittsburgh, Pennsylvania. PPT
- Dialogue Models. September 18, 2003. Graduate Seminar
on Dialog Processing. Pittsburgh, Pennsylvania. PPT
- Engineering Dialog for Gadgets. September 12, 2003. 1st Annual LTI Student
Research Symposium. Pittsburgh, Pennsylvania. PPT
- The Universal Speech Interface. June 3, 2003. Pittsburgh
Digital Greenhouse Progress Report. Pittsburgh, Pennsylvania. PPT
- James: A Personal Mobile Universal Speech Interface for Electronic
Devices. November 22, 2002. Dialog on Dialogs Reading
Group. Pittsburgh, Pennsylvania. PPT
- From Grammars to N-grams: Estimating N-grams From a
Context-Free Grammar and Sparse Data. May 16, 2002. Sphinx
Lunch. Pittsburgh, Pennsylvania. PPT
- XML Output for Sphinx. June 22, 2001. Sphinx
Lunch. Pittsburgh, Pennsylvania. PPT
Software
- TeamTalk
is a multi-participant multi-modal human-robot interface.
- Olympus
is a research framwork for spoken language interfaces.
- cfg2ngram is a program for combining
knowledge from a context-free grammar and sparse (or no) data to
produce a tri-gram language model.
- decodereval is a program for evaluating a grammar and a
speech decoder using raw audio utterances and oracle
transcriptions. It produces word error rates and concept error
rates.
- sphinxptk is a graphical front-end
for Sphinx.
This page created and maintained by Thomas Harris.
Last updated on Wednesday August 01, 2007.