Yuan-Shan Lee

Ph.D., Department of Computer Science & Information Engineering, National Central University

CV (PDF)

Email: kg934283@gmail.com
Fields of Interest: Machine Learning, Deep Learning, Signal Processing, Multimedia Processing

Selected Journal Articles

J.-C. Wang, Y.-S. Lee, C.-H Lin, S.-F. Wang, C.-H. Shih, and C.-H. Wu, “Compressive Sensing-Based Speech Enhancement,” IEEE Transactions on Audio Speech and Language Processing, vol. 24, no. 11, pp. 2122-2131, 2016. (SCI, Rank factor: 4/31=13%)

J.-C. Wang, Y.-S. Lee, Y.-H. Chin, Y.-R. Chen, and W.-C. Hsieh, “Hierarchical Dirichlet Process Mixture Model for Music Emotion Recognition,” IEEE Transactions on Affective Computing, vol. 6, no. 3, pp. 261-271, 2015. (SCI, Rank factor: 19/123=15%)

J.-C. Wang, Y.-S. Lee, C.-H. Lin, E. Siahaan, and C.-H. Yang, “Robust Environmental Sound Recognition Using Fast Subspace Based Noise Suppression,” IEEE Transactions on Automation Science and Engineering, vol. 12, no. 4, pp. 1235-1242, 2015. (SCI, Rank factor: 11/58=18%)

Y.-S. Lee, Y.-L. Chiang, P.-R. Lin, T.-C. Tai, and C.-H. Lin, “Robust and Efficient Content-based Music Retrieval System,” APSIPA Transactions on Signal and Information Processing, vol. 5, pp. 1-9, 2016. (Invited submission)

Selected Conference Papers

S.-H. Chen, Y.-S. Lee, and J.-C. Wang , “Locality Preserving Discriminative Complex-valued Latent Variable Model ,” IEEE ICPR 2018, Beijing, China , Aug. 2018. (Accepted)

S.-H. Wu, Y.-S. Lee, S.-H. Chen, and J.-C. Wang , “Learning a Hierarchical Latent Semantic Model for Multimedia Data,” IEEE ICPR 2018, Beijing, China , Aug. 2018. (Accepted)

S.-H. Chen*, Y.-S. Lee*, M.-C. Hsieh, and J.-C. Wang , “Playing Technique Classification Based on Deep Collaborative Learning of Variational Auto-Encoder And Gaussian Process,” IEEE ICME 2018, San Diego, USA , Jul. 2018. (* indicates the co-first author)

S.-H. Chen*, Y.-S. Lee*, and J.-C. Wang , “Complex-valued Gaussian Process Latent Variable Model for Phase-Incorporating Speech Enhancement,” IEEE ICASSP 2018, Alberta, Canada, Apr. 2018. (* indicates the co-first author)

S.-H. Chen, Y.-S. Lee, Y.-S. Hsu, C.-H. Wu, and J.-C. Wang , “Locality-Preserving Complex-Valued Gaussian Process Latent Variable Model for Robust Face Recognition,” IEEE ICASSP 2018, Alberta, Canada, Apr. 2018.

Y.-S. Lee, K. Yu, S.-H. Chen, and J.-C Wang, “Discriminative Training of Complex-valued Deep Recurrent Neural Network for Singing Voice Separation,” ACM Multimedia 2017, Mountain View, CA USA, Oct. 2017.

S.-H. Chen, S.-H. Wu, Y.-S. Lee, and J.-C Wang, “Hierarchical Representation based on Bayesian Nonparametric Tree-Structured Mixture Model for Playing Technique Classification,” ACM Multimedia Thematic Workshops 2017, CA USA, Oct. 2017.

Y.-S. Lee, C.-Y. Wang, S.-F. Wang, J.-C. Wang, and C.-H. Wu, “Fully Complex Deep Neural Network for Phase-Incorporating Monaural Source Separation,” IEEE ICASSP 2017, New Orleans, USA, Mar. 2017, pp. 281-285.

V.-H Duong, Y.-S. Lee, J.-J. Ding, B.-T. Pham, M.-Q. Bui, P. T. Bao, and J.-C. Wang, “Exemplar-Embed Complex Matrix Factorization for Facial Expression Recognition,” IEEE ICASSP 2017, New Orleans, USA, Mar. 2017, pp. 1837-1841.

S.-H. Chen*, A. Hernawan*, Y.-S. Lee*, and J.-C. Wang, “Hand Gesture recognition based on Bayesian sensing hidden Markov models and Bhattacharyya divergence," IEEE ICIP 2017, Beijing, China, Sep. 2017. (* indicates the co-first author)

T. Pham, Y.-S. Lee, S. Mathulaprangsan, and J.- C. Wang, “Source Separation Using Dictionary Learning and Deep Recurrent Neural Network with Locality Preserving Constraint,” IEEE ICME 2017, Hong Kong, Jul. 2017.

Y.-S. Lee, C.-Y. Wang, S. Mathulaprangsan, J.-H. Zhao, and J.-C. Wang, “Locality Preserving K-SVD via Joint Dictionary and Classifier Learning for Object Recognition,” ACM Multimedia 2016, Amsterdam, The Netherlands, Oct. 2016, pp.481-485.

T. Pham, Y.-S. Lee, Y.-B. Lin, T.-C. Tai, and J.-C. Wang, “Single Channel Source Separation Using Sparse NMF and Graph Regularization,” ASE International Conference on Social Informatics, Oct. 2015. (Best Paper Award)

Honors

Gold Medal Award, Merry Electronics Co., (2017)

Ph.D. Dissertation Award, IICM, (2017)

Ph.D. Dissertation Award, ACLCLP, (2017)

Ph.D. Dissertation Award, TAAI, (2017)

Ph.D. Dissertation Award, WIC, (2017)

IEEE ICASSP Signal Processing Society (SPS) Travel Grant, (2017)

Presidential Award-Highest honor for postgraduate, NCU, (2016)

Best Paper Award, ASE International Conference on Social Informatics, (2015)

Excellent Student Award, NCU, (2015)

Honorary Member of Phi Tau Phi Scholastic Honor Society, (2013)

Research

I am developing methods for analyzing, extracting, recognizing, and retrieving information from multimedia signals, with special emphasis on speech, image, and music/audio signals. My research interests include, but are not limited to: source separation, speech enhancement, object recognition, music emotion recognition, sound event recognition, deep neural network, and Bayesian learning.

1. Speech Processing

Research in the field of speech processing has focused on speech enhancement and source separation. The goal of speech enhancement and source separation is to remove interference from a noisy speech recording. Recent work includes a novel compressive sensing-based method for speech enhancement, a phase-incorporating complex-valued deep neural network for source separation, and a locality-constrained recurrent neural network (RNN) for singing source separation. Ongoing research includes a complex-valued deep RNN for source separation, a CS-based underdetermined source separation, and a complex-valued Gaussian process (GP) for speech enhancement.

2. Image Processing

With respect to image processing, researches has focused on object recognition and face recognition. Object recognition is a hot topic in the field of computer vision that involves recognizing objects from photographs or videos. Recent work includes a locality-preserving dictionary learning method for object recognition and a complex-valued dictionary learning method for facial recognition. Ongoing research includes a locality-preserving complex-valued Gaussian process latent variable model (GPLVM) for face recognition and a Bayesian sensing hidden Markov models (BS-HMM) based hand gesture recognition system.

3. Music/Audio Processing

Research in the field of music/audio processing has focused on the recognition of emotion in music, music retrieval, and sound event recognition. Recent achievement in this field is the development of a robust environmental sound recognition that and a novel multi-label music emotion recognition (MER) system. Ongoing research includes a kernelized joint dictionary method that involves collaborative representation, a complex-valued GPLVM for sound event recognition, and a playing technique classification that is based on hierarchical latent Dirichlet allocation (h-LDA).