Jilt Sebastian

Short Biography


Jilt Sebastian is a pre-doctoral fellow in the audio signal processing group at Idiap Research Institute, Martigny, Switzerland affiliated to Swiss Federal Institute of Technology at Lausanne (EPFL), and also a fifth year Direct Ph.D student (Masters by Research + Ph.D) in the Department of Computer Science and Engineering at Indian Institute of Technology Madras. Previously, received B Tech (Hons.) from Mahatma Gandhi University, Kerala in 2011 in Electronics and Communication Engg.

His current interests mainly include statistical pattern classification, signal processing, artificial neural networks with applications to audio source separation, pitch estimation, musical onset detection, spike estimation from neuronal signals and, speaker diarization.

Skills


  • Programming Languages: Python, perl, shell script, C++
  • Scientific Computing Tools:Tensorflow, MATLAB, Kaldi toolkit, HTK toolkit, Audacity, Sonic visualizer, SAP ERP Toolkit

Work and Education


  • IDIAP Research Institute, Martigny, Switzerland (2017 - Present) , Pred-doc, Audio signal processing
  • Indian Institute of Technology, Madras (2013 - present) , Doctoral student in Computer Science and Engineering Dept., Speech Signal Processing
  • Signal Analysis and Interpretation Lab (SAIL), University of Southern California, United States (2017 March-May) , Visiting Scholar, Speaker Diarization group
  • Department of Brain and Cognitive Sciences, Massechussets Institute of Technology (MIT), Cambridge, United States (2017 March) , Visiting Scholar, Sur Lab
  • Mahatma Gandhi University, Kerala (2007 - 2011) , B. Tech (Hons. ) in Electronics and Communication Engineering

Selected Publications


  • An analysis of the high resolution property of group delay function with applications to audio signal processing, J Sebastian, M Kumar, HA Murthy, Speech Communication 81, 42-53, 2016
  • Group delay based music source separation using deep recurrent neural networks, J Sebastian, HA Murthy, Signal Processing and Communications (SPCOM), 2016 International Conference on,
  • GDspike: An accurate spike estimation algorithm from noisy calcium fluorescence signals, J Sebastian, YS Sreekar, RV Rikhye, M Sur, HA Murthy, Acoustics, Speech and Signal Processing (ICASSP), 2017 IEEE International Conference on, 2017.
  • Onset Detection in Composition Items of Carnatic Music, J Sebastian, HA Murthy, International Symposium on Music Information Retrieval (ISMIR), 2017.
  • For the full list of publications: Google Scholar

Awards and Honours


  • Swiss Government Excellence fellowship for Foreign Scholars 2017-2018 , to conduct research at Idiap affiliated to EPFL
  • MHRD Fellowship, During Ph.D (2013-2018).
  • Best student award, Amal Jyothi College of Engineering, 2011,



Tel: +41 27 721 7705
Office: 303-6
Contact

jilt.sebastian@idiap.ch