| Discriminant independent component analysis CS Dhir, SY Lee IEEE transactions on neural networks 22 (6), 845-857, 2011 | 51 | 2011 |
| Efficient feature selection based on information gain criterion for face recognition CS Dhir, N Iqbal, SY Lee 2007 International Conference on Information Acquisition, 523-527, 2007 | 43 | 2007 |
| Unsupervised style and content separation by minimizing mutual information for speech synthesis TY Hu, A Shrivastava, O Tuzel, C Dhir ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and …, 2020 | 38 | 2020 |
| A filter bank approach to independent component analysis for convolved mixtures HM Park, CS Dhir, SH Oh, SY Lee Neurocomputing 69 (16-18), 2065-2077, 2006 | 32 | 2006 |
| Apparatus and method for recognizing content using audio signal H Cho, J Lee, CS Dhir US Patent 8,886,635, 2014 | 30 | 2014 |
| Stacked 1D convolutional networks for end-to-end small footprint voice trigger detection T Higuchi, M Ghasemzadeh, K You, C Dhir arXiv preprint arXiv:2008.03405, 2020 | 23 | 2020 |
| Hybrid transformer/ctc networks for hardware efficient voice triggering S Adya, V Garg, S Sigtia, P Simha, C Dhir arXiv preprint arXiv:2008.02323, 2020 | 22 | 2020 |
| Optimize what matters: Training dnn-hmm keyword spotting model using end metric A Shrivastava, A Kundu, C Dhir, D Naik, O Tuzel ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and …, 2021 | 19 | 2021 |
| Extraction of independent discriminant features for data with asymmetric distribution CS Dhir, J Lee, SY Lee Knowledge and information systems 30, 359-375, 2012 | 19 | 2012 |
| Streaming transformer for hardware efficient voice trigger detection and false trigger mitigation V Garg, W Chang, S Sigtia, S Adya, P Simha, P Dighe, C Dhir arXiv preprint arXiv:2105.06598, 2021 | 12 | 2021 |
| Dynamic curriculum learning via data parameters for noise robust keyword spotting T Higuchi, S Saxena, M Souden, TD Tran, M Delfarah, C Dhir ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and …, 2021 | 11 | 2021 |
| Hessian-free optimization for learning deep multidimensional recurrent neural networks M Cho, C Dhir, J Lee Advances in Neural Information Processing Systems 28, 2015 | 10 | 2015 |
| Hybrid feature selection: Combining fisher criterion and mutual information for efficient feature selection CS Dhir, SY Lee International Conference on Neural Information Processing, 613-620, 2008 | 10 | 2008 |
| Multi-task learning with cross attention for keyword spotting T Higuchil, A Gupta, C Dhir 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU …, 2021 | 9 | 2021 |
| Device-directed speech detection: Regularization via distillation for weakly-supervised models V Garg, O Rudovic, P Dighe, AH Abdelaziz, E Marchi, S Adya, C Dhir, ... arXiv preprint arXiv:2203.15975, 2022 | 8 | 2022 |
| Filterbank-based blind signal separation with estimated sound direction HM Park, CS Dhir, DK Oh, SY Lee 2005 IEEE International Symposium on Circuits and Systems (ISCAS), 5874-5877, 2005 | 6 | 2005 |
| Permutation correction of filter bank ICA using static channel characteristics CS Dhir, HM Park, SY Lee International Conference on Neural Information Processing, 1076-1081, 2004 | 6 | 2004 |
| Blind source separation of noisy observations using directionally constrained Filterbank ICA SY Lee, CS Dhir, HM Pa, DK Oh International Conference on Neural Information Processing, 0-0, 2005 | 4 | 2005 |
| Traffic lane recognizing apparatus and method thereof Y Park, J Kim, J Lee, H Kim, J Park, LEE Jeihun, A Park, CS Dhir US Patent App. 14/236,099, 2014 | 3 | 2014 |
| Information-theoretic feature extraction and selection for robust classification CS Dhir, SY Lee Independent Component Analyses, Wavelets, Neural Networks, Biosystems, and …, 2009 | 2 | 2009 |
Soo-Young LeeProfessor of Electrical Engineering, KAISTVerified email at kaist.ac.kr
