People > Faculty
Katrin Kirchhoff
Katrin Kirchhoff Research Assistant Professor Signal and Image Processing M422 EE/CSE Box 352500 University of Washington Seattle, WA 98195 |
| Phone: (206) 616-5494 E-mail: katrin@ee.washington.edu |
University of Bielefeld, Germany PhD 1999
University of Bielefeld, Germany MA 1996
Biosketch
Dr. Kirchhoff joined the Speech, Signal and Language Interpretation (SSLI) Lab in the EE Department at the University of Washington in November 1999, after receiving her PhD in Computer Science from the University of Bielefeld, Germany. Most of here previous work has been in the area of automatic speech recognition. In her dissertation work, part of which was performed as a visitor at the International Computer Science Institute in Berkeley, California, she developed articulatory-based models for automatic speech recognition. Currently she is working on various projects in speech and natural language processing, including automatic language and speaker identification, dialogue systems and automatic speech summarization.
Research Interests
Speech recognition, language and speaker identification, multilingual speech processing, interfaces between speech recognition and natural language processing, dialogue act detection, machine learning, classifier combination.
Research Projects
- Communicator
Automatic detection and repairs of communication failures in automated dialogue systems. - Automatic Speech Summarization
Use of speech recognition to automatically transcribe, index and summarize recordings of lectures and talks. - Feature-based Language Identification
Techniques for automatically identifying the language of a speaker based on statistical models of articulatory configurations estimated from the acoustic signal. - Speech Recognition and Human Speech Perception
Development and comparison of computational models for human speech perception and automatic speech recognition.
For more information, go to my SSLI lab homepage: http://ssli.ee.washington.edu/ssli/people/katrin/
Selected Publications
- K. Kirchhoff and S. Parandekar, "Multi-stream statistical N-gram modeling with application to automatic language identification", Proceedings of Eurospeech 2001, Aalborg, Denmark, September 2001
- K. Kirchhoff, "A Comparison of Classification Techniques for the Automatic Detection of Error Corrections in Human-Computer Dialogues", Proceedings of the NAACL Workshop on Adaptation in Dialogue Systems, Pittsburgh, PA, June 2001
- K. Kirchhoff and J. Bilmes. "Combination and Joint Training of Acoustic Classifiers for Speech Recognition", Proceedings of ASR2000, Paris, France, 2000
- K. Kirchhoff, G.A. Fink and G. Sagerer, "Conversational Speech Recognition Using Acoustic and Articulatory Input", ICASSP-2000, Istanbul, Turkey, June 2000
- K. Kirchhoff. "Combining Articulatory and Acoustic Information for Speech Recognition in Noisy and Reverberant Environments", Proceedings of the International Conference on Spoken Language Processing, Sydney, Australia, December, 1998, 891-894
- Ganapathiraju A, Goel V, Picone J, Corrada A, Doddington G, Kirchhoff K, Ordowski M and Wheatley B. "Syllable - A Promising Recognition Unit for LVCSR", Proceedings of the IEEE Automatic Speech Recognition and Understanding Workshop, Santa Barbara, USA, December 1997
