William Hartmann
email: hartmann dot 59 at osu dot edu
I am a sixth year PhD student in the CSE Department at OSU.
I am working with Professor Eric Fosler-Lussier in the Speech and Technologies Lab.
Curriculum Vitae
Research

I am broadly interested in artificial intelligence and machine learning. More specifically, I am interested in robust automatic speech recognition. I am currently investigating methods for estimating binary masks using top-down linguistic information and methods for incorporating those masks into automatic speech recognition.
Publications

W. Hartmann and E. Fosler-Lussier (2012). "ASR-Driven Binary Mask Estimation using Spectral Priors", in the Proceedings of ICASSP, Kyoto. (pdf)

W. Hartmann and E. Fosler-Lussier (2011). "Investigations into the incorporation of the ideal binary mask in ASR", in the Proceedings of ICASSP, Prague. (pdf)

R. Prabhavalkar, P. Jyothi, W. Hartmann, J. Morris, and E. Fosler-Lussier (2010). "Investigations into the Crandem Approach to Word Recognition", In the Proceedings of NAACL-HLT, Los Angeles, pp. 725-728. (pdf) (bibtex)

W. Hartmann and E. Fosler-Lussier (2009). "Investigating phonetic information reduction and lexical confusability", in the Proceedings of Interspeech, Brighton, UK., pp. 1659-1662. (pdf) (bibtex)

R. Fox and W. Hartmann (2009). "Using Context to Improve Hand-Written Character Recognition", in the International Society for Advanced Science and Technology Transactions on Computers and Intelligent Systems, No. 1, Vol. 1, pp. 40-49.

R. Fox and W. Hartmann (2006). "An Abductive Approach to Hand-written Character Recognition for Multiple Domains", in The Proceedings of the 2006 International Conference on Artificial Intelligence, Volume II, pp. 349-355, H. Arabnia editor, CSREA Press.

R. Fox and W. Hartmann (2005). "Hand-written Character Recognition Using Layered Abduction", in Advances in Systems, Computing Sciences and Software Engineering, the Proceedings of SCSS, T. Sobh and K. Elleithy editors, pp. 141-147, Springer.
Manuscripts in Preparation
Listed journal and conference names are the expected venues for submission.

W. Hartmann, A. Narayanan, E. Fosler-Lussier, and D.L. Wang. "Nothing Doing: Re-evaluating Missing Data ASR", IEEE Transactions on Audio, Speech, and Language Processing.

W. Hartmann and E. Fosler-Lussier. "Improved Model Selection for the ASR-Driven Binary Mask", Interspeech.

W. Hartmann and E. Fosler-Lussier. "Underspecified Pronunciation Models for Automatic Speech Recognition", Speech Communication.
Courses