The Ph.D. Dissertation.

Slides for Defense.

Title and Abstract pdf file .

Acknowledgement pdf file .

Table of Contents, pdf file .

List of Figures, pdf file .

LIst of Tables, pdf file .

Chapter 1, Introduction, pdf file .

Chapter 2, Maximum Entropy Modeling, pdf file .

Chapter 3, Hierarchical Training Methods, pdf file .

Chapter 4, Topic Dependent Language Models, pdf file .

Chapter 5, Syntactic Language Models, pdf file .

Chapter 6, Composite Language Model with Both Topic Dependencies and Syntactic Dependencies, pdf file .

Chapter 7, Improving the Computation in Using ME Models, pdf file .

Chapter 8, Conclusion, pdf file .

Appendix A, Sign Test (in ps) in PDF .

Appendix B, Design Document (in ps) in PDF .

Appendix C, User Manual (in ps) in PDF .

Bibliography.

ps file for the whole dissertation, revised on 10/10/2002.

pdf file for the whole dissertation, revised on 10/10/2002.

Selected Publications and Technical Reports.

Jun Wu and Sanjeev Khudanpur,Building A Topic-Dependent Maximum Entropy Language Model for Very Large Corpora, In Proceedings of ICASSP2002, Orlando, USA, May 12-17, 2002.

Woosung Kim, Sanjeev Khudanpur and Jun Wu, Smoothing Issues in the Structured Language Model, Proceedings of Eurospeech 2001, pp. 717-720, September 2000, Denmark.

Jun Wu and Sanjeev Khudanpur, Efficient Training Methods for Maximum Entropy Language Modeling, Proceedings of ICSLP2000, Vol. 3, pp. 114-117, Oct. 2000, Beijing, China.

Sanjeev Khudanpur and Jun Wu, Maximum Entropy Techniques for Exploiting Syntactic, Semantic and Collocational Dependencies in Language Modeling, Computer Speech and Language, pp. 355-372, Oct. 2000.

Jun Wu and Sanjeev Khudanpur, Syntactic Heads in Statistical Language Modeling. Proceedings of ICASSP2000, pp. 1699-1702, June 4-9, Istanbul, Turkey.

Jun Wu and Sanjeev Khudanpur, Combining Nonlocal, Syntactic and N-Gram Dependencies in Language Modeling. Proceedings of Eurospeech'99, vol 5, pp2179-2182, September 6-10, 1999, Budapest, Hungary. Winner Eurospeech99 ESLNET Best Student Paper Award.

Sanjeev Khudanpur and Jun Wu, A Maximum Entropy Language Model to Integrate N-Grams and Topic Dependencies for Conversational Speech Recognition. Proceedings of ICASSP'99, pp. 553-556, March 14-19, 1999, Phoenix.

Eric Brill and Jun Wu, Classifier Combination for Improved Lexical Disambiguation , Processings of COLING-ACL'98, pp 191-195, Auguest 10-14, 1998, Montreal Canada.

Jun Wu and Zuoying Wang, Entropy of Chinese and the Perplexity of the Language Models, ACTA Electronica Sinica, v24 n10 Oct 1996, p69-71.(EI)

Jun Wu, Zuoying Wang and Jiasong Sun, A Method of Speech Understanding for Chinese Speech Recognition, Chinese Software Journal, Special Issue, Oct 1996, pp188-193

Jun Wu, et al., Entropy for Chinese Text, Chinese Information Journal, Vol. 6 N.2, pp. 17-23, 1996.

Zuoying Wang, Jun Wu (actual writer), et al., Methods Towards the Very Large Vocabulary Chinese Speech Recognition, Proceedings of Eurospeech' 95, pp.215-218, 1995.9, Madrid, Spain.

Jun Wu, Z. Wang, Y. Feng, Automatic Classification of Chinese Texts, Chinese Information Journal, Vol. 5 N.4, pp. 34-41, 1995.

Jun Wu, Z. Wang, J. Sun, J. Guo, Chinese Speech Understanding and Spelling-Word Translation Based on the Statistics of Corpus, ICSLP'94 , September 18-22, Yokohama, Japan. (SCI)

Jun Wu, Z. Wang, Y. Ren, Stochastic Language Models for Chinese Speech Recognition Based on Chinese Spelling., 1994 International Symposium on Speech, Image Processing and Neural Networks, pp. 674-677., April 1994, HongKong. (SCI)

Technical Reports

Jun Wu, Maximum Entropy Language Modeling with Non-Local Dependencies, Dissertation Proposal.

Jun Wu, Using Multiple Taggers to Improve English Part-of-Speech Tagging by Error Learning, Qualification Project Report of the Johns Hopkins University.

Jun Wu, A Maximum Entropy Language Model with Topic Sensitive Features, Qualification Project Report of the Johns Hopkins University.

Translation Works

Jun Wu, Jiasong Sun, Huizhong Yang and Yan Zhang, IBM OS/2 Program Guide (From English Edition to Chinese Edition), Tsinghua Press, 1994.

Other Publications.

Jun Wu, et al., Large Vocabulary Telephone Speech Recognition System, The 4th National Conference on Man-Machine Speech Communication, 1996, 10, Beijing China.

Jun Wu, Z. Wang, et al., A Strategy for Large Vocabulary Continuous Speech Recognition, Proc. of ICCC, Singapore, June, 1996.

Jun Wu, Z. Wang, J. Yu, Research on the Perplexity of Language Models, The 2nd National Conference on Intelligent Computer Interface and Applications of Artificial Intelligence, pp. 169-174, July, 1995, Weihai, China.

Jun Wu, Z. Wang, J. Sun, J. Guo, An Approach toward Speech Understanding and Spelling-word Translation, The 3rd National Conference on Man-Machine Speech Communication, pp. 214 -220, 1994.10, Chongqing, China.

Zuoying Wang, Jun Wu, et al., JANET - A New Chinese Speech Recognition, Understanding, and Synthesis System, The 3rd National Conference on Man-Machine Speech Communication, pp.482-486, 1994.10, Chongqing, China.

Jun Wu, Z. Wang, Y. Ren, A New Approach to Speech Understanding Based on Corpus, The 2nd National Conference of Computer Linguistics, pp. 96-101, Nov. 1993. Xiamen China.

Jun Wu, Z. Wang, Stochastic Language Models for Eliminating Chinese Speech Recognition Error, The 1st National Conference of Intelligent Computer Interface and Application of Artificial Intelligence, pp. 233-238, July 22-24, 1993, Jinbo Lake, China.

Last updated by Jun Wu, Jun 2002.