Wooil Kim

Ph.D.

Research Assistant Professor
Center for Robust Speech Systems
Dept. of Electrical Engineering
Erik Jonsson School of Engineering and Computer Science
University of Texas at Dallas


Office
Phone
Email
Homepage


ECSN 4.322
(972) 883-4388, (972) 795-0065(cp)
wikim{at}utdallas.edu
www.utdallas.edu/~wikim

Research Interests
    1. Information Retrieval System
      - Multimedia/Music Information Retrieval, Spoken Document Retrieval
       - Keyword Spotting/Detection, Data Mining, Dialog System
    2. Human Behavior Signal Processing
      - In-Vehicle Human Interface, Robust Speech Recognition for In-Vehicle System
       - Signal Processing/Estimation Under Cognitive Stress/Distraction
    3. Speech Processing/Recognition
      - Large Vocabulary Continuous Speech Recognition, ASR for Wireless/VoIP
       - Environmental Robust Processing for ASR, Missing-Feature Theory
    4. Statistical Signal Processing, Pattern Recognition, Neural Network

Recent and Current Works

    1. SpeechFind system (SpeechFind.utdallas.edu)
      - Spoken document retrieval system of National Gallery of the Spoken Word
       - System integration and programming
       - Algorithm development for environmental robustness
     2. Feature compensation method based on Gaussian mixture model
      - Aiming at robust speech recognition in additive background noise
       - Adaptive estimation of parameters using PCMM(Parallel Combined Mixture Model) scheme
       - Computational reduction via multiple-model interpolation and mixture-sharing
       - Discriminability improvement applying MCE(Minimum Classification Error) training
     3. Mask estimation for missing-feature algorithm
      - Aiming at robust speech recognition in unknown noisy environments
       - Mask estimation based on Bayesian classifier
       - Environment-independent performance via model training using the colored-noise

Education and Experiences

    1. Experiences
    2. Education
      Ph.D., Electronics Engineering, Korea University, Seoul, Korea, Aug. 2003
        -Thesis: Model-based Feature Compensation for Robust Speech Recognition in Adverse Environments
         -Advised by Prof. Hanseok Ko
      M.S., Electronics Engineering, Korea University, Seoul, Korea, Aug. 1998
        -Thesis: Spectral Subtraction based on Phonemic and Auditory Properties
         -Advised by Prof. Hanseok Ko
      B.S., Electronics Engineering, Korea University, Seoul, Korea, Feb. 1996

Publications

    1. Journal Papers
      [1] Wooil Kim and J.H.L. Hansen, "Missing-Feature Reconstruction by Leveraging Temporal Spectral Correlation for Robust Speech Recognition in Background Noise Conditions," IEEE Transactions on Audio, Speech, and Language Processing, 2010.
       [2] Wooil Kim and J.H.L. Hansen, "Phonetic Distance Based Confidence Measure," IEEE Signal Processing Letters, vol.17, no.2, pp.117-120, Feb. 2010.
       [3] Wooil Kim and J.H.L. Hansen, "Time-Frequency Correlation Based Missing-Feature Reconstruction for Robust Speech Recognition in Band-Restricted Conditions," IEEE Transactions on Audio, Speech, and Language Processing, vol.17, no.7, pp.1292-1304, Sept. 2009.
       [4] Wooil Kim and J.H.L. Hansen, "Feature Compensation in the Cepstral Domain Employing Model Combination," Speech Communication, vol.51, no.2, pp.83-96, Feb. 2009.
       [5] Wooil Kim and J.H.L. Hansen, "Feature Compensation Employing Multiple Environmental Models for Robust In-Vehicle Speech Recognition," IEICE Trans. Information and Systems, Vol. E91-D, No. 3, pp. 430-438, March 2008.
       [6] Wooil Kim and H. Ko, "Noise Variance Estimation for Kalman Filtering of Noisy Speech," IEICE Trans. Information and Systems, Vol. E84-D, No. 1, pp. 155-160, Jan, 2001.
       [7] Wooil Kim, S. Kang, and H. Ko, "Spectral subtraction based on phonetic dependency and masking effects," IEE Proc.-Vision Image and Signal Processing, Vol. 147, No. 5 , pp.423-427, Oct. 2000.
     2. Conference Papers (Peer-Reviewed)
      [1] Wooil Kim and J. H. L. Hansen, "Angry Emotion Detection from Real-Life Conversational Speech by Leveraging Content Structure," ICASSP-2010, Dallas, U.S.A., March 2010.
       [2] Wooil Kim and J. H. L. Hansen, "Mask Estimation Employing Posterior-Based Representative Mean for Missing-Feature Speech Recognition with Time-Varying Background Noise," IEEE ASRU-2009, pp. 194-198, Merano, Italy, Dec. 2009.
       [3] Wooil Kim and J. H. L. Hansen, "Variational Model Composition for Robust Speech Recognition with Time-Varying Background Noise," Interspeech-2009, pp. 2399-2402, Brighton, UK, Sept. 2009.
       [4] Wooil Kim and J. H. L. Hansen, "Robust Angry Speech Detection Employing TEO-Based Discriminative Classifier Combination," Interspeech-2009, pp. 2019-2022, Brighton, UK, Sept. 2009.
       [5] Wooil Kim and J. H. L. Hansen, "Missing-Feature Method for Speaker Recognition in Band-Restricted Conditions," Interspeech-2008, pp.1909-1912, Brisbane, Australia, Sept. 2008.
       [6] J.H.L. Hansen, Wooil Kim, and P. Angkititrakul, "Advances in Human-Machine Systems for In-Vehicle Environments," IEEE HSCMA-2008: Hands-free Speech Communication and Microphone Arrays, pp. 128-131, Trento, Italy, May 2008.
       [7] Wooil Kim and J. H. L. Hansen, "Advances in Spoken Document Retrieval for the U. S. Collaborative Digitization Program," IEEE ASRU-2007, pp.687-692, Kyoto, Japan, Dec. 2007.
       [8] Wooil Kim and J. H. L. Hansen, "Advances in SpeechFind: Transcript Reliability Estimation Employing Confidence Measure based on Discriminative Sub-word Model for SDR," Interspeech-2007, pp.2409-2412, Antwerp, Belgium, Aug. 2007.
       [9] Wooil Kim, M. Akbacak and J. H. L. Hansen, "Advances in SpeechFind: CRSS-UTD Spoken Document Retrieval System," ACM SIGIR 2007 Workshop, Amsterdam, Netherlands, July 2007.
       [10] Wooil Kim and J. H. L. Hansen, "Missing-Feature Reconstruction for Band-Limited Speech Recognition in Spoken Document Retrieval," Interspeech-2006, pp.2306-2309, Pittsburgh, U.S.A., Sep. 2006.
       [11] Wooil Kim and R. M. Stern, "Band-Independent Mask Estimation for Missing-Feature Reconstruction in the Presence of Unknown Background Noise," ICASSP-2006, pp.305-308, Toulouse, France, May 2006.
       [12] Wooil Kim, R. M. Stern and H. Ko, "Environment-Independent Mask Estimation for Missing-Feature Reconstruction," Interspeech-2005, pp.2637-2640, Lisbon, Portugal, Sep. 2005.
       [13] Wooil Kim, O. Kwon and H. Ko, "PCMM-based Feature Compensation Schemes Using Model Interpolation and Mixture Sharing," ICASSP-2004, pp.989-992, Montreal, Canada, May 2004.
       [14] Wooil Kim, S. Ahn and H. Ko, "Feature Compensation Scheme Based on Parallel Combined Mixture Model," Eurospeech-2003, pp. 677-680, Geneva, Switherland, Sep. 2003.
       [15] Wooil Kim and H. Ko, "Improved Acoustic Modeling Based on Selective Data-Driven PMC," ICASSP-2002, Student Forum, Orlando, U.S.A., May 2002.
       [16] Wooil Kim, T. Kim, S. Ahn and H. Ko, "Model Based Stress Decision Method," Eurospeech-2001, Vol. 1, pp.107-110, Alborg, Denmark, Sep. 2001.
    3. Book Chapter
      [1] Wooil Kim and J. H. L. Hansen, "SpeechFind: Advances in Rich Content Based Spoken Document Retrieval," Chapter 17 of Handbook of Research on Digital Libraries: Design, Development, and Impact, pp.173-187, IGI Global, 2009.
       [2] Wooil Kim and J. H. L. Hansen, "Feature Compensation Employing Model Combination for Robust Speech Recognition in In-Vehicle Environment," Chapter 19 of In-Vehicle Corpus and Signal Processing for Driver Behavior, pp.233-243, Springer, 2008.
     4. Etc.