Resume: Pedro J. Moreno

this resume needs some updating, but for the time being it willhave to do...

Research Interests

  • Speech Indexing and recognition.

  • Machine learning and its application to multimedia (audio, images, video)

Work Experience
  • Research Scientist . Google Inc., New York, NY

  • Senior Member of Technical Staff, Project Leader, HP Labs, Cambridge Research Laboratory. July 2002 - April 2004. Led research projects in genomics, biosignal (cardiology), audio, image, text and multimedia indexing and machine learning methods. Developed the core technology of NewsTuner a radio similarity search engine.

  • Senior Member of Technical Staff, Project Leader, Compaq Labs, Cambridge Research Laboratory. 2000 - June 2002. Led research projects in media indexing, information retrieval, information fusion, machine learning. Design and implementation of audio processing aspects of distributed system for content-based audio processing and indexing (www.speechbot.com). This was the first audio search engine based on speech recognition technology publicly available on the web. Led discussions with venture capitalist and explore potential spinoff ideas for COMPAQ.

  • Member of the Research Staff Digital Equipment Corporation, Cambridge Research Laboratory. July 1996 - 2000. Research projects in speech recognition, speaker identification and noise robusness.

  • Research Assistant, Electrical and Computer Engineering Department, Carnegie Mellon University,  Sep. 1991 - May. 1996. Design and implementation of algorithms for speech recognition in noisy environments. Lead CMU participation in several DARPA supported evaluations of speech recognition in noisy and telephone environments. Documented and created tools for acoustic modeling for the CMU sphinx large vocabulary speech recogntion system.

  • Visiting Researcher, Bell Laboratories, Murray Hill, NJ January 1989-August 1991. Worked with David Roe, Fernando Pereira and other senior researchers on speech to speech translation systems. Our system (VEST) was continuosly demonstrated at the World Fair in Seville in 1992. Also conducted research in noise robustness.

Education
  • Ph.D., Electrical and Computer Engineering, Carnegie Mellon University, 1996. Thesis Title: Speech Recognition in Noisy Environments.

  • M.S., Electrical and Computer Engineering, Carnegie Mellon University, 1992.

  • M.S., Ingeniero Superior de Telecomunicaciones, Universidad Politecnica de Madrid, Madrid, SPAIN, 1988.

Awards
  • Fulbright Fellowship to conduct graduate studies in the U.S.A. 1991-1995

Recent Professional Activities
  • Member of Technical Committee for ICSLP 2006
  • Reviewer for Eurospeech 2005, ICASSP 2006, ICML 2006, ICSLP 2006
  • Reviewer for ICSLP 2004/2/0, Eurospeech 2004-1, ICML 2004
  • Reviewer for Speech Communication, Journal of Machine Learning
  • Reviewer for IEEE Speech and Signal Procession, IEEE Multimedia, Signal Processing Letters.
Pre-Google Patents
  • Cardiac Diagnostic System and Method, United States Patent Application 20050222508, Pedro J. Moreno, David Goddeau, Beth T. Logan.
  • Computer Method and system for reading and analyzing ECG signals, United States Patent Application 20050222507, Beth T. Logan, Pedro J. Moreno, David Goddeau.
  • Method and Apparatur for object identification, classification or verification, , United States Patent Application 20050044053, Pedro J. Moreno, Purdy Ho.
  • Computer Method and Apparatus for Uniform Representation of Genome Sequences, United States Patent Application 200304450, Simon Kasif, Beth Logan, Pedro Moreno, Baris Suzek
  • Method to expand inputs for word or document searching, United States Patent Application 20030187649, Beth T. Logan, JM Van Thong, Pedro J. Moreno
  • Vocabulary independent speech decoder system and method using subword units, United States Patent Application  20030187643, JM Van Thong, Pedro J. Moreno, Edward Whitakker
  • Method for refining time alignments of closed captions, United States Patent, 6,442,518, J. M. Van Thong, Pedro J. Moreno
  • Computer method and apparatus for segmenting text streams, United States Patent 6,772,120, Pedro J. Moreno, David M. Blei
  • System and method for detecting repetitions in a multimedia stream,  United States Patent Application 20030101144, Pedro J. Moreno
  • Method for refining time alignments of closed captions, United States Patent 6,442,518, JM Van Thong, Pedro J. Moreno
  • Environmently compensated speech processing, United States Patent 5,924,065, Brian S. Eberman, Pedro J. Moreno

Publications

there is a bunch of HP technical reports out there. But they are mostly copies of the papers listed below.

Peered reviewed conference papers

  • SVM kernel adaptation in speaker classification and verification, Pedro J. Moreno, Purdy Ho, International Conference on Speech and Language Processing (ICSLP-2004), October 2004, Jeju (Korea).

  • The Kullback-Leibler Kernel as a Framework for Discriminant and Localized Representations for Visual Recognition N. Vasconcelos, P. Ho, and P. Moreno, Proceedings of the European Conference on Computer Vision, Prague, Czech, 2004.

  • A Kullback-Leibler Divergence Based Kernel for SVM Classification in Multimedia Applications, Pedro J. Moreno, Purdy Ho, Nuno Vasconcelos. NIPS-2003, December, 2003, Vancouver.

  • A New SVM Approach to Speaker Identification and Verification using Probabilistic Distance Kernels, Pedro J. Moreno, Purdy Ho, Proceedings of the 2003 Eurospeech Conference, Geneva, Switzerland, Sept. 2003.

  • A Family of Probabilistic Kernels based on Information Divergence, A. B. Chan, N. Vasconcelos, Pedro J. Moreno. Workshop on Graphical Models and Kernels, NIPS 2004.

  • An Experimental Study of EM-based algorithms for semi-supervised learning in audio classification, Pedro J. Moreno and Shivani Agarwal. Workshop on Unlabeled data, COLT, 2003.

  • Fusion of Semantic and Acoustic Approaches for Spoken Document Retrieval, Beth Logan, Patrawadee Prasangsit and Pedro Moreno, ISCA Workshop on Multilingual Spoken Document Retrieval (MSDR 2003), April 2003. Appears as Technical Report HPL-2003-55.

  • Topic Segmentation with an Aspect Hidden Markov Model, David Blei and Pedro J. Moreno, Proceedings of the 2001 SIGIR Conference, New Orleans, LA, September 2001.

  • Remote Homology Detection Using Feature Vectors Formed Using Alignments of Small Motifs, Beth Logan, Pedro Moreno, Baris Suzek, Zhiping Weng, and Simon Kasif, 6-th Annual International Conference on Computational Molecular Biology (RECOMB), April 2002.

  • Word and Sub-word Indexing Approaches for Reducing the Effects of OOV Queries on Spoken Audio, Beth Logan, Pedro Moreno and Om Deshmukh, Human Language Technology Conference (HLT), March 2002.

  • A Boosting Approach for Confidence Scoring, Pedro Moreno, Beth Logan and Bhiksha Raj, 7th European Conference on Speech Communication and Technology (Eurospeech), September 2001.

  • SpeechBot: a Content-based Search Index for Multimedia on the Web, Pedro Moreno, JM Van Thong, Beth Logan, Blair Fidler, Katrina Maffey, and Matthew Moores. First IEEE Pacific-Rim Conference on Multimedia, (IEEE-PCM 2000), 2000.

  • An Experimental Study of an Audio Indexing System for the Web, Beth Logan, Pedro Moreno, Jean-Manuel Van Thong and Ed Whittaker, International Conference on Speech and Language Processing, 2000.

  • SpeechBot: a Speech Recognition based Audio Indexing System for the Web, David Goddeau, Anna Litvinova, Beth Logan, Pedro Moreno, Michael Swain, and Jean-Manuel Van Thong, RIAO'2000.

  • Factorial HMMs for Acoustic Modeling, B.T.Logan and P.J.Moreno, Proceedings IEEE International Conference on Acoustics, Speech and Signal Processing, 1998.

  • Compensation for Environmental Degradation in Automatic Speech Recognition, R. M. Stern, B. Raj, and P. J. Moreno, (1997). Proc. of the ESCA Tutorial and Research Workshop on Robust Speech Recognition for Unknown Communication Channels, April, 1997, Pont-au-Mousson, France, pp. 33-42

  • A Vector Taylor Series Approach For Environment-Independent Speech Recognition, P. J. Moreno, B. Raj, and R. M. Stern, Proc. of the ICASSP, Atlanta, GA, May 1996.

  • Cepstral Compensation By Polynomial Approximation For Environment-Independent Speech Recognition, B. Raj, E. Gouvêa, P. J. Moreno, and R. M. Stern, Proc. of the ICSLP, Philadelphia, PA, Oct. 1996.

  • Adaptation and Compensation: Approaches To Microphone And Speaker Independence In Automatic Speech Recognition, E. B. Gouvea, P. J. Moreno, B. Raj, T. M. Sullivan, and R. M. Stern, Proceedings of the ARPA Workshop on Speech Recognition Technology, Harriman, NY, Morgan Kaufmann, D. Pallett, Ed.

  • Recognition Of Continuous Broadcast News With Multiple Unknown Speakers And Environments, U. Jain, M. A. Siegler, S.-J. Doh, E. Gouvea, P. J. Moreno, B. Raj, and R. M. Stern, Proceedings of the ARPA Workshop on Speech Recognition Technology, Harriman, NY, Morgan Kaufmann, D. Pallett, Ed.

  • Multivariate-Gaussian-Based Cepstral Normalization for Robust Speech Recognition, P. J. Moreno, B. Raj, E. Gouvêa, and R. M. Stern, Proc. of the ICASSP, Detroit, Michigan, 1995.

  • A Unified Approach to Robust Speech Recognition, P. J. Moreno, B. Raj, R. M. Stern, Proc. of Eurospeech-95, Madrid, Spain, September, 1995.

  • Continuous Speech Recognition of Large Vocabulary Telephone Quality Speech, P. J. Moreno, M. A. Siegler, U. Jain, and R. M. Stern, Proc. of the Eighth Spoken Language Systems Technology Workshop, 1995.

  • Approaches to Microphone Independence in Automatic Speech Recognition, P. J. Moreno, U. Jain, B. Raj, and R. M. Stern, Proc. of the Eighth Spoken Language Systems Technology Workshop, 1995.

  • Approaches to Environment Compensation in Automatic Speech Recognition, P. J. Moreno, B. Raj, and R. M. Stern, Proc. 15th International Conference on Acoustics, Trondheim, Norway, Vol. III, pp. 109-112, June, 1995.

  • Sources of Degradation of Speech Recognition in the Telephone Network, P. J. Moreno, and R. M. Stern, Proc. of the ICASSP, Adelaide, Australia, 1994.

  • Signal Processing for Robust Speech Recognition, R. M. Stern, F.-H. Liu, P. J. Moreno, and A. Acero, Proc. of the International Conference on Spoken Language Processing, Yokohama, Japan, September, 1994.

  • Efficient Grammar Processing For a Spoken Language Translation System, David B. Roe, Pedro J. Moreno, Richard W. Sproat, Fernando Pereira, Michael Riley, and Alejandro Macarrón. In Proc. of ICASSP 1992

Journal papers

  • From Multimedia Retrieval to Knowledge Management, Pedro J. Moreno, Jean Manuel Van Thong, Beth Logan and Gareth Jones, IEEE Computer Magazine, April 2002. (Invited paper).

  • SpeechBot: An Experimental Speech-based Search Engine for Multimedia Content on the Web, Jean Manuel Van Thong, Pedro J. Moreno, Beth Logan, Blair Fidler, Katrina Maffey and Matthew Moores, IEEE Transactions on Multimedia, Vol 4, Nr. 1, March 2002.

  • Data Driven Environmental Compensation for Speech Recognition: A Unified View. Pedro J. Moreno, Bhiksha Raj, R.M. Stern, Speech Communication, 24, 267-285, 1998

  • A spoken language translator for restricted-domain context-free languages: David B. Roe, Pedro J. Moreno, Richard W. Sproat, Fernando Pereira, Speech Communication 1992

Book Chapters

  • Probabilistic semantic image annotation and retrieval. Nuno Vasconcelos, Gustavo Carneiro, Antoni Chan, Pedro J. Moreno. upcoming in 2006...