I currently work on a couple of projects specifically at Beckman. With Mark Hasegawa-Johnson and his student Po-Sen Huang I collaborate on speech enhancement approaches that utilize deep learning. The goal of this work is to produce a system that can remove interference from speech recorded in noisy environments. This has resulted in a series of papers and also a conference award. With Art Kramer and his student Aki Nikolaidis I’m working on statistical analysis of brain data as
collected from certain task experiments. This investigation is geared towards understanding plasticity and how various areas of the brain get utilized. We have one publication out of this work and Aki has also secured some funding on it (and will be a Beckman Institute Graduate Fellow in 2015-2016).
- Traa, J.; Wingate, D.; Stein, N. D.; Smaragdis, P., Robust Source Localization and Enhancement with a Probabilistic Steered Response Power Model. IEEE-ACM Transactions on Audio Speech and Language Processing 2016, 24, (3), 493-503.
Nikolaidis, A.; Goatz, D.; Smaragdis, P.; and Kramer, A. F., Predicting Skill-Based Task Performance and Learning with fMRI Motor and Subcortical Network Connectivity, in Proceedings of the Fifth International Workshop on Pattern Recognition in Neuroimaging, 2015, Stanford University.
- Huang, P. S.; Kim, M.; Hasegawa-Johnson, M.; Smaragdis, P., Joint Optimization of Masks and Deep Recurrent Neural Networks for Monaural Source Separation. IEEE-ACM Transactions on Audio Speech and Language Processing 2015, 23, (12), 2136-2147.
Smaragdis, P.; Fevotte, C.; Mysore, G. J.; Mohammadiha, N.; Hoffman, M., Static and Dynamic Source Separation Using Nonnegative Factorizations a Unifed View. IEEE Signal Processing Magazine 2014, 31, (3), 66-75, DOI: 10.1109/msp.2013.2297715.
Huang P.-S.; Kim, M.; Hasegawa-Johnson, M.; Smaragdis, P., Singing-Voice Separation from Monaural Recordings Using Deep Recurrent Neural Networks, in Proceedings of the International Symposium of Music Information Retrieval, 2014. Taipei, Taiwan.
Huang, P. S.; Kim, M.; Hasegawa-Johnson, M. A.; Smaragdis, P., Deep Learning for Monaural Speech Separation, Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2014, Florence, Italy.
Kim, M.; Smaragdis, P., Collaborative Audio Enhancement Using Probabilistic Latent Component Sharing, In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, New York, 2013, 896-900.
Kim, M.; Smaragdis, P., Manifold Preserving Hierarchical Topic Models for Quantization and Approximation, International Conference on Machine Learning, Atlanta, Georgia, June 2013.
Kim, M.; Smaragdis, P., Single Channel Source Separation Using Smooth Nonnegative Matrix Factorization with Markov Random Fields, IEEE Workshop for Machine Learning in Signal Processing, Southampton, United Kingdom, September 2013.
Mohammadiha, N.; Smaragdis, P.; Leijon, A. Low-Artifact Source Separation Using Probabilistic Latent Component, IEEE Workshop for Applications of Signal Processing in Audio and Acoustics, New Paltz, New York, October 2013.
Mohammadiha, N.; Smaragdis, P.; Leijon, A. Simultaneous Noise Classification and Reduction Using a Priori Learned Models, IEEE Workshop for Machine Learning in Signal Processing, Southampton, United Kingdom, September 2013.
Mohammadiha, N.; Smaragdis, P.; Leijon, A., Supervised and Unsupervised Speech Enhancement Using Nonnegative Matrix Factorization. IEEE Transactions on Audio Speech and Language Processing 2013, 21, (10), 2140-2151, DOI: 10.1109/tasl.2013.2270369.
Pathak, M. A.; Raj, B.; Rane, S.; Smaragdis, P., Privacy-Preserving Speech Processing. IEEE Signal Processing Magazine 2013, 30, (2), 62-74.
Smaragdis, P.; Kim, M. Non-Negative Matrix Factorization for Irregularly-Spaced Transforms, IEEE Workshop for Applications of Signal Processing in Audio and Acoustics, New Paltz, New York, October 2013.
Traa, J.; Smaragdis, P., A Wrapped Kalman Filter for Azimuthal Speaker Tracking. IEEE Signal Processing Letters 2013, 20, (12), 1257-1260, DOI: 10.1109/lsp.2013.2287125.
Traa, J.; Smaragdis, P.; Blind Multi-Channel Source Separation by Circular-Linear Statistical Modeling of Phase Differences, In 2013 IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, New York, 2013, 4320-4324.
Smaragdis, P.; Raj, B., The Markov Selection Model for Concurrent Speech Recognition. Neurocomputing 2012, 80, 64-72.
The Communications Office maintains the information included in Beckman Institute's online directory listings. In order to update your directory listing, please submit the following information to firstname.lastname@example.org:
- a short bio including information on your educational background and your field
- any honors and awards you may have received
- a description of your research (approximately 200-400 words)
- a list of recent representative publications
- a photo of yourself (you can submit one or we can take one for you)