The Neural Systems Analysis Laboratory, directed by John C. Malone assistant professor Archana Venkataraman in the Department of Electrical and Computer Engineering, recently had eight different papers accepted to academic conferences.
The papers were penned by five doctoral candidates – Ravi Shankar, Jeff Craley, Sayan Ghosal, Niharika Shimona D’Souza, and Naresh Nandakumar – and were accepted to two different conferences. Four papers got into Interspeech 2019: Conference of the International Speech Communication Association, while another four were accepted to MICCAI: International Conference on Medical Image Computing and Computer Assisted Intervention.
“What makes this situation unique is the breadth of our research, from epileptic seizure localization, to fusing imaging and genetic data, to morphing emotional cues in human speech,” Venkataraman said. “I think it shows that we are tackling really important scientific problems in novel ways.”
The papers accepted to Interspeech, which runs from September 15 through 19 in Graz, Austria, are particularly exciting for Venkataraman because they focus on emotion morphing in human speech, which is a new area of research for the laboratory.
The project is based around human speech being a carrier of the speaker’s intention and emotion, but the decrypting of emotions in speech is still not fully understood. The group’s goal is to identify those features of speech, which characterize emotion and computationally manipulate these components to change emotional inflections, with the four papers being broadly based around attempting to solve this problem through different approaches.
Shankar wrote the four papers focusing on that topic, two of which have been selected for oral presentations at Interspeech. It’s an experience that Shankar expects will play a critical role in his research going forward.
“I am looking forward to it because it’s for the first time that I will be presenting our work to people outside of my lab,” Shankar said. “I’m hoping to get some constructive criticism from the audience so that we can further improve the models that we have worked on so far. For us PhD students, we can expand our knowledge and understanding by either reading papers published by other groups or by discussing with our peers. Participation in this international conference belongs to the latter category.”
Below is the list of papers from the Neural Systems Analysis Laboratory that were recently accepted to conferences:
R. Shankar, J. Sager and A. Venkataraman. A Multi-Speaker Emotion Morphing Model Using Highway Networks and Maximum Likelihood Objective. To Appear in Interspeech: Conference of the International Speech Communication Association, 2019. Selected for an Oral Presentation
J. Sager, R. Shankar, J. Reinhold, and A. Venkataraman. VESUS: A Crowd-Annotated Database to Study Emotion Production and Perception in Spoken English. To Appear in Interspeech: Conference of the International Speech Communication Association, 2019. Selected for an Oral Presentation
R. Shankar, J. Sager, H.-W. Hsieh, N. Charon and A. Venkataraman. Automated Emotion Morphing in Speech Based on Diffeomorphic Curve Registration and Highway Networks. To Appear in Interspeech: Conference of the International Speech Communication Association, 2019.
R. Shankar and A. Venkataraman. Weakly Supervised Syllable Segmentation by Vowel-Consonant Peak Classification. To Appear in Interspeech: Conference of the International Speech Communication Association, 2019.
J. Craley, C. Jouny, E. Johnson and A. Venkataraman. Automated Noninvasive Seizure Detection and Localization Using Switching Markov Models and Convolutional Neural Networks. To Appear in MICCAI: International Conference on Medical Image Computing and Computer Assisted Intervention, 2019. Selected for Early Acceptance
S. Ghosal, Q. Chen, A.L. Goldman, W. Ulrich, K.F. Berman, D.R. Weinberger, V.S. Mattay and A. Venkataraman. Bridging Imaging, Genetics, and Diagnosis in a Coupled Low-Dimensional Framework. To Appear in MICCAI: International Conference on Medical Image Computing and Computer Assisted Intervention, 2019. Selected for Early Acceptance
N.S. D’Souza, N. Wymbs, M.B. Nebel, S. Mostofsky and A.~Venkataraman. Integrating Neural Networks and Dictionary Learning for Multidimensional Clinical Characterizations from Functional Connectomics Data. To Appear in MICCAI: Medical Imaging Computing and Computer Assisted Intervention, 2019.
N. Nandakumar, K. Manzoor, J.J. Pillai, S.K. Gujar, H.I. Sair, and A. Venkataraman. A Novel Graph Neural Network to Localize Eloquent Cortex in Brain Tumor Patients from Resting-State fMRI Connectivity. To Appear in MICCAI: Medical Imaging Computing and Computer Assisted Intervention Connectomics in Neuroimaging (CNI) Workshop, 2019.
-by Wick Eisenberg