Sanjeev P. Khudanpur

Assistant Professor

and Member

E mail:
Telephone:
Fax:

Office Location:



Office Hours:
Department of Electrical and Computer Engineering &
Department of Computer Science (secondary appointment)
Center for Language and Speech Processing

My last-name at jhu dot edu
+1.410.516.7024
+1.410.516.5050

325 Coumptational Sciences and Engineering Building
3400 North Charles Street
Baltimore, MD 21218

By appointment (Tue-Thu 10:30-12:00)

Biosketch: I received the B.Tech. degree in Electrical Engineering from the Indian Institute of Technology, Bombay, in 1988, and the Ph.D. degree in Electrical and Computer Engineering from the University of Maryland, College Park, in 1997. My doctoral dissertation was supervised by Prof. Prakash Narayan and was titled Model Selection and Universal Data Compression.

Since 1996, I have been on the faculty of the Johns Hopkins University, serving until June 2001 as Associate Research Scientist in the Center for Language and Speech Processing and, since then, as Assistant Professor in the Department of Electrical and Computer Engineering and the Department of Computer Science.

In Fall 2000, I held a visiting appointment in the Institute for Mathematics and its Applications (IMA), University of Minnesota, Minneapolis, MN. I organized two IMA workshops on the role of mathematics in multimedia - ``Mathematical Foundations of Speech Processing and Recognition,'' and ``Mathematical Foundations of Natural Language Modeling.''

Click here for a full vita or a two-page NSF-style resume.


Teaching: 520.447 Introduction to Information Theory and Coding (Fall semesters)
520.651 Random Signal Analysis (Fall semesters)
520.666 Information Extraction from Speech and Text (Spring semesters)
520.674 Information Theoretic Methods in Statistics (Spring semesters)


Research: I am interested in the application of information theoretic methods to human language technologies such as automatic speech recognition, machine translation and natural language processing. All these technologies make heavy use of statistical models of human language. I am interested in understanding the structure of such models and in estimating their parameters from data.

I also organize the CLSP Summer Workshops to advance the greater research agenda of this field.


Publications in Peer-Reviewed Journals:

A. Kanlis, S. Khudanpur and P. Narayan, "Typicality of a Good Rate-Distortion Code," in Problemy Peredachi Informatsii (Problems of Information Transmission), 32(1):96-103, 1996. (click here for the English version)

M. Riley, W. Byrne, M. Finke, S. Khudanpur, A. Ljolje, J. McDonough, H. Nock, M. Saraclar, C. Wooters and G. Zavaliagkos, "Stochastic Pronunciation Modeling from Hand-Labelled Phonetic Corpora," in Speech Communication, 29(2-4):209-224, 1999.

M. Saraclar, H. Nock and S. Khudanpur, "Pronunciation Modeling by Sharing Gaussian Densities Across Phonetic Models," in Computer Speech and Language, 14(2):137-160, 2000.

S. Khudanpur and J. Wu, "Maximum Entropy Techniques for Exploiting Syntactic, Semantic and Collocational Dependencies in Language Modeling," in Computer Speech and Language, 14(4):355-372, 2000.

S. Khudanpur and P. Narayan, "Order Estimation for a Special Class of Hidden Markov Sources and Binary Renewal Processes," in IEEE Transactions on Information Theory, 48(6):1704-1713, 2002.

D. He, D. Oard, J. Wang, J. Luo, D. Demner-Fushman, K. Darwish, P. Resnik, S. Khudanpur, M. Nossal and A. Leuski, "Making MIRACLEs: Interactive Translingual Search for Cebuano and Hindi," in ACM Transactions on Asian Language Information Processing, 2(3):219-244, 2003.

S. Khudanpur and W. Kim, "Contemporaneous Text as Side Information in Statistical Language Modeling," in Computer Speech and Language, 18(2):143-162, 2004.

H. Meng, B. Chen, S. Khudanpur, G. Levow, W. Lo, D. Oard, P. Schone, K. Tang, H. Wang and J. Wang, "Mandarin-English Information (MEI): Investigating Translingual Speech Retrieval," in Computer Speech and Language, 18(2):163-179, 2004.

W. Kim and S. Khudanpur, "Lexical Triggers and Latent Semantic Analysis for Cross-Lingual Language Model Adaptation ," in ACM Transactions on Asian Language Information Processing, 3(2):94-112, 2004.

M. Saraclar and S. Khudanpur, "Pronunciation Change in Conversational Speech and Its Implications for Automatic Speech Recognition," in Computer Speech and Language, 18(4):375-395, 2004.

B. Jedynak and S. Khudanpur, "Maximum Likelihood Set for Estimating a Probabilty Mass Function," in Neural Computation, 17(7):1508-1530, 2005.

D. Karakos, S. Khudanpur, D. Marchette, A. Papamarcou and C. Priebe, "On the Minimization of Concave Information Functionals for Unsupervised Classification via Decision Trees," in Statistics and Probability Letters, 2007 (to appear).

Publications in Refereed Conference Proceedings:

These are listed in my full vita.


Edited Books: Mathematical Foundations of Speech and Language Processing,
M. Johnson, S. Khudanpur, M. Ostendorf and R. Rosenfeld (Editors),
IMA Volumes in Mathematics and Its Applications, Vol. 138, Springer-Verlag, New York, Jan 2004. (Table of Contents)


Colleagues: Faculty
Bill Byrne, ECE
Jason Eisner, CS
Fred Jelinek, ECE
Don Geman, AMS
Damianos Karakos, CLSP
Michael Miller, BME
David Yarowsky, CS

Former Advisees (Degree) and where they went after graduation ...
Sourin Das (MSE) Constellation Energy Inc.
Woosung Kim (PhD), Convergys Inc.
Srividya Mohan (MSE), Multimodal Technologies Inc.
Srihari Reddy (MSE), Yahoo! Inc.
Murat Saraclar (PhD), Bogazici University
Paola Virga (MSE), IBM T. J. Watson Research Center
Jun Wu (PhD), Google Inc.

Current Advisees
Arnab Ghoshal, ECE
Ali Yazgan, CS

Administrative Staff
Laura Graham, CLSP
Eiwe Lingefors, CLSP
Sue Porterfield, CLSP


Personal: I spent three fun-filled weeks in Kenya (and Tanzania) with two of my nieces and their parents - my sister and brother-in-law - in May and June of 2002. We visited lots of national parks and game reserves and took lots of pictures with a simple point-and-shoot camera. Here are a few of them in a 52MB Powerpoint document or a 310MB PDF document. Take a peek if you have a few minutes (or a few hours, depending on your modem-speed!).



Send comments about this page to Sanjeev Khudanpur
Last modified: Mon Mar 3 19:20:06 EST 2008