Ciprian Chelba: Resume

Research Interests

• Statistical Modeling applied to Natural Language and Speech
• Machine Learning
• Information Theory and Statistics

Work Experience and Accomplishments

February 2006 - present:	Research Scientist, Google, Mountain View, CA
2000-2006:	Researcher, Microsoft Research, Redmond, WA
	Projects: Keyword Search in Collections of Spoken Documents, collaboration with MIT (TJ Hazen, Jim Glass) Discriminative Language Modeling for Conversational Speech, collaboration with Patrick Nguyen Speech and Text Routing, How May I Help You (HMIHY), collaboration with Milind Mahajan Hoolie Spelling Mnemonics Language Model, collaboration with David Mowatt (product team); -to ship in Windows Vista Automatic Design of Question Sets for HMM State Clustering, collaboration with Rachel Morton (product team) Automatic Capitalization of Uniform Case Text
1995-2000:	Research Assistant, Center for Speech and Language Processing, The Johns Hopkins University
	Projects: Developed a language model that makes use of syntactic structure (see Publications) SLM code release
July-August 1996:	Researcher, Center for Speech and Language Processing, The Johns Hopkins University
	Projects: DoD Workshop on Innovative Techniques for LVCSR at The Johns Hopkins University Dependency Language Modeling group (see Publications)
June-August 1995:	Summer Intern, I.B.M., T.J. Watson Research Center
	Projects: Speech Recognition Group: Phone length modeling for Continuous Speech Recognition

Education

• The Johns Hopkins University, Baltimore, MD
  Ph.D. in Electrical and Computer Engineering, January 2000; Advisor: Frederick Jelinek
  Ph.D. thesis: Exploiting Syntactic Structure for Natural Language Modeling
  M.S. in Electrical Engineering, May 1996
• "Politehnica" University, Bucharest, Romania
  Faculty of Electronics and Telecommunications
  Diploma Engineer in Electrical Engineering, Sept 1993
  Diploma thesis: Neural Controller for BUCK Converter; Matlab simulation (see AWARDS)

Awards

• Computer Speech and Languge Best Paper Award (2002)
• TEMPUS (European Community) 6-month scholarship (Feb 1993 -- Aug 1993) for developing the Diploma Thesis at Politecnico di Torino, Italy.
• ACL-EACL'97 Student Grant.

Community

• Speech and Language TC seat on the HLT Advisory Board, 2007
• Member of the advisory board to the Professional Masters Program in Computational Linguistics at the
  University of Washington, 2006
  
• Invited (and accepted) to join IEEE Speech and Language Technical Committee, area of Speech Mining, Summarization and Document Retrieval, 2006 (expires 2008)
  
• Member of Publicity Committee for HLT-NAACL'06 conference, area speech processing
• Panelist at HLT-NAACL'06 Doctoral Consortium
• Invited to participate in the NSF sponsored JHU/Center for Language and Speech Processing workshop, 2005
• NSF panel reviewer: IT4IISLL (Division of Information and Intelligent Systems, Language and Linguistics) panel headed by Mary Harper, 2004
• Program committee member for ACL05 demo session
• Program committee member for workshop `Incremental Parsing: Bringing together NLP and Cognitive Modeling', EMNLP 2004
• Program committee member for workshop `Spoken Language Understanding', EMNLP 2004
• Chaired Language Modeling and Identification session, ICASSP 2001
• Reviewer for Computer Speech and Language, Computational Linguistics, Speech Communication, IEEE Transactions for Speech and Audio Processing, IEEE Signal Processing Magazine
• Reviewer for ACL, HLT, HLT-NAACL, EMNLP, COLING, Eurospeech, ICASSP, NIPS, AAAI and satellite workshops

Professional Affiliation

• IEEE Senior Member
• ACL

Publications

C. Chelba
A Structured Language Model Proceedings of ACL/EACL, Madrid, Spain, 1997, Student Session( browse slides)
C. Chelba et al.
Structure and Performance of a Dependency Language Model Proceedings of Eurospeech, Rhodes, Greece, 1997
C. Chelba and F. Jelinek
Exploiting Syntactic Structure for Language Modeling Proceedings of COLING-ACL, Montreal, Canada, 1998( browse slides)
C. Chelba and F. Jelinek
Refinement of a Structured Language Model Proceedings of ICAPR, Plymouth, UK, 1998
C. Chelba and F. Jelinek
Structured Language Modeling for Speech Recognition (extended project abstract) Proceedings of NLDB, Klagenfurt, Austria, 1999
C. Chelba and F. Jelinek
ERRATA to Structured Language Modeling for Speech Recognition (extended project abstract)
C. Chelba and F. Jelinek
Recognition Performance of a Structured Language Model Proceedings of Eurospeech, Budapest, Hungary, 1999
F. Jelinek and C. Chelba
Putting Language into Language Modeling Proceedings of Eurospeech, Budapest, Hungary, 1999
C. Chelba and F. Jelinek
Structured Language Modeling Computer Speech and Language (2000), 14 (2002 Best Paper award)
C. Chelba
Portability of Syntactic Structure for Language Modeling Proceedings of ICASSP, Salt Lake City, UT, 2001
C. Chelba and M. Mahajan
Information Extraction using the Structured Language Model Proceedings of EMNLP, Pittsburgh, PA, 2001 ( browse slides)
C. Chelba and P. Xu
Richer Syntactic Dependencies for Structured Language Modeling Proceedings of ASRU, Madona di Campiglio, Italy, 2001 ( poster)
P. Xu, C. Chelba and F. Jelinek
A Study on Richer Syntactic Dependencies for Structured Language Modeling Proceedings of ACL, Philadelphia, PA, 2002
C. Chelba and R. Morton
Mutual Information Phone Clustering for Decision Tree Induction Proceedings of ICSLP, Denver, CO, 2002
Y. Wang et al.
Combination of Statistical and Rule-based Approaches for Spoken Language Understading Proceedings of ICSLP, Denver, CO, 2002
C. Chelba, M. Mahajan, A. Acero
Speech Utterance Classification Proceedings of ICASSP, Hong Kong, 2003
C. Chelba and A. Acero
Discriminative Training of N-gram Classifiers for Speech and Text Routing Proceedings of Eurospeech, Geneva, Switzerland, 2003 ( poster)
C. Chelba and A. Acero
Conditional ML Estimation Using Rational Function Growth Transform Snowbird Learning Workshop, Utah, April 2004 ( poster)
J. G. Kahn, M. Ostendorf and C. Chelba
Parsing Conversational Speech Using Enhanced Segmentation Proceedings of HLT/NAACL, Boston, MA, 2004
C. Chelba and A. Acero
Adaptation of Maximum Entropy Capitalizer: Little Data Can Help a Lot Proceedings of EMNLP, Barcelona, Spain, 2004 (Powerpoint slides)
C. Chelba and A. Acero
Adaptation of maximum entropy capitalizer: Little data can help a lot Computer Speech and Language (2005), Volume 20, Issue 4
C. Chelba and A. Acero
Position Specific Posterior Lattices for Indexing Speech Proceedings of ACL, Ann Arbor, 2005
C. Chelba and A. Acero
SPEECH OGLE: Indexing Uncertainty for Spoken Document Search Proceedings of ACL Demo Poster Session, Ann Arbor, 2005 ( poster)
C. Chelba and A. Acero
Indexing Uncertainty for Spoken Document Search Proceedings of Eurospeech, Lisbon, Portugal, 2005 (slides, Internet Explorer only *mht format)
J. Silva, C. Chelba, A. Acero
Pruning Analysis of the Position Specific Posterior Lattices for Spoken Document Search Proceedings of ICASSP, Toulouse, 2006 (poster)
C. Chelba
Acoustic Sensitive Language Model Perplexity for Automatic Speech Recognition Snowbird Learning Workshop, Utah, April 2006 ( poster)
Zhou et al.
Towards Spoken-Document Retrieval for the Internet: Lattice Indexing For Large-Scale Web-Search Architectures Proceedings of the Human Language Technology Conference of the NAACL, Main Conference, NYC, June 2006
J. Silva, C. Chelba, A. Acero
Integration of Metadata in Spoken Document Search Using Position Specific Posterior Lattices IEEE Spoken Language Technology Workshop, Aruba 2006( poster)
C. Chelba, J. Silva and A. Acero
Soft Indexing of Speech Content for Search in Spoken Documents Computer Speech and Language, Volume 21, Issue 3, July 2007, Pages 458-478

Technical reports

C. Chelba and M. Mahajan
Growth Transform for Conditional Maximum Likelihood Estimation of Log-Linear Models, MSR-TR-2002-65, Microsoft Research, June 2002

C. Chelba and A. Acero
Conditional Maximum Likelihood Estimation of Naive Bayes Probability Models Using Rational Function Growth Transform, MSR-TR-2004-33, Microsoft Research, April 2004

Tutorials

C. Chelba and T. J. Hazen
Automatic Spoken Document Processing for Retrieval and Browsing HLT-NAACL, NYC, June 2006

C. Chelba and T. J. Hazen
Automatic Spoken Document Processing for Retrieval and Browsing, ICASSP, Honolulu, April 2007

Invited Talks

C. Chelba
Exploiting Syntactic Structure for Natural Language Modeling LSUW Colloquium Series, University of Washington, October 02
C. Chelba
Human-Computer Interaction on Relational Domains Using Natural Language CLSP Seminar Series, The Johns Hopkins University, March 03
C. Chelba
Exploiting Syntactic Structure for Natural Language Modeling LSUW Colloquium Series, University of Washington, January 2005
C. Chelba
Indexing Uncertainty for Efficient Search in Spoken Documents (Internet Explorer only *mht format) EE SSLI Seminar Series, University of Washington, November 2005
C. Chelba
Spoken Document Retrieval and Browsing Vilem Mathesius Lecture Series, Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Prague, December 2006
C. Chelba
Spoken Document Retrieval and Browsing CLSP Seminar Series, The Johns Hopkins University, February 2007
C. Chelba
Spoken Document Retrieval and Browsing University of Washington, July 2007

Patent Applications

Filed as co-inventor for 15 US patents among which 7 have been filed internationally

Miscellaneous Information

• Romanian citizen, current visa status: US permanent resident
• Foreign languages: Italian, Romanian