Deepak Baby

Personal Webpage

Short Biography

Deepak Baby is currently working as a post-doctoral researcher with Prof. Hervé Bourlard at Idiap Research Institute since May 2019. His main research is focussed on sparse and hierarchical representations for speech modelling. He previously worked as a post-doctoral researcher with Prof. Sarah Verhulst at WAVES, Department of Information Technology, Ghent University between February 2017 and April 2019, where he was working on DNN-based hearing restoration strategies for hearing impairment.

Deepak Baby received his PhD in Electrical Engineering (Thesis title : Non-negative Sparse Representations for Speech Enhancement and Recognition) from ESAT/PSI group in KU Leuven in November 2016 under the supervision of Prof. Hugo Van hamme. He received the Masters degree in Communication Engineering from Dept. of Electrical Engineering, Indian Institute of Technology Bombay in 2012 and Bachelors degree in Electronics and Communication Engineering from College of Engineering Trivandrum, India in 2009.


Work Experience

  • Postdoctoral Researcher, Speech and Audio Processing Group, Idiap Research Institute, Martigny, Switzerland. (May 2019 - Present)
  • Postdoctoral Researcher, WAVES Research Group, Ghent University, Belgium. (Feb 2017 - Apr 2019)
  • Visiting Researcher, Nuance Communications Inc., Merelbeke, Belgium. (Apr - Jun 2015)
  • Visiting Researcher, Tampere Univsersity of Technology, Finland. (Jun - Aug 2013)

Education

  • PhD in Electrical Engineering, KU Leuven, Belgium. (Jul 2012- Nov 2016)
    Thesis title: Non-negative sparse representations for speech enhancement and recognition
  • MTech in Communication Engineering, Indian Institute of Technology Bombay, India. (Jul 2010 - Jun 2012)
    Thesis title: Extensions to greedy algorithms in compressed sensing
  • BTech in Electronics and Communication Engineering, College of Engineering Trivandrum, Kerala, India. (Aug 2005 - Aug 2009)

Publications

Google Scholar

Journals

  1. Deepak Baby and Hugo Van hamme. Joint denoising and dereverberation using exemplarbased sparse representations and decaying norm constraint. IEEE/ACM Trans. Audio, Speech and Language Processing, 25(10):2024–2035, 2017.
  2. Deepak Baby, Tuomas Virtanen, Jort F. Gemmeke, and Hugo Van hamme. Coupled dictionaries for exemplar-based speech enhancement and automatic speech recognition. IEEE/ACM Trans. Audio, Speech and Language Processing, 23(11):1788–1799, 2015.

Conferences

  1. Fotios Drakopoulos, Deepak Baby, and Sarah Verhulst. Real-Time Audio Processing on a Raspberry Pi using Deep Neural Networks. In 23rd International Congress on Acoustics(ICA), Aachen, Germany, June 2019.
  2. Deepak Baby and Sarah Verhulst. SERGAN: Speech enhancement using relativistic generative adversarial networks with gradient penalty. In ”Acoustics, Speech and Signal Processing (ICASSP), 2016 IEEE International Conference on, Brighton, UK, May 2019.
  3. Deepak Baby and Sarah Verhulst. Biophysically-inspired features improve the generalizability of neural network-based speech enhancement systems. In Proc. INTERSPEECH, pages 3264–3268, Hyderabad, India, September 2018. ISCA.
  4. Deepak Baby and Hugo Van hamme. Supervised Speech Dereverberation in Noisy Environments using Exemplar-based Sparse Representations. In Acoustics, Speech and Signal Processing (ICASSP), 2016 IEEE International Conference on, pages 156–160, Shanghai, China, March 2016.
  5. Deepak Baby and Hugo Van hamme. Hybrid Input Spaces for Exemplar-based Noise Robust Speech Recognition using coupled dictionaries. In 23rd European Signal Processing Conference (EUSIPCO), pages 1676–1680, Nice, France, September 2015.
  6. Deepak Baby and Hugo Van hamme. Investigating Modulation Spectrogram Features for Deep Neural Network-based Automatic Speech Recognition. In Proc. INTERSPEECH, pages 2479–2483, Dresden, Germany, September 2015.
  7. Emre Yilmaz, Deepak Baby, and Hugo Van hamme. Noise Robust Exemplar Matching for Speech Enhancement: Applications to Automatic Speech Recognition. In Proc. INTERSPEECH, pages 688–692, Dresden, Germany, September 2015.
  8. Emre Yilmaz, Deepak Baby, and Hugo Van hamme. Noise Robust Exemplar Matching with Coupled Dictionaries for Single-Channel Speech Enhancement. In 23rd European Signal Processing Conference (EUSIPCO), pages 874–878, Nice, France, September 2015.
  9. Deepak Baby, Jort F. Gemmeke, Tuomas Virtanen, and Hugo Van hamme. Exemplar-based Speech Enhancement for Deep Neural Network based Automatic Speech Recognition. In Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on, pages 4485–4489, Brisbane, Australia, April 2015.
  10. Deepak Baby, Tuomas Virtanen, Jort F. Gemmeke, Tom Barker, and Hugo Van hamme. Exemplar-based Noise Robust Speech Recognition using Modulation Spectrogram Features. In Spoken Language Technology Workshop (SLT), 2014 IEEE, pages 519–524, South Lake Tahoe, USA, December 2014.
  11. Deepak Baby, Tuomas Virtanen, Tom Barker, and Hugo Van hamme. Coupled Dictionary Training for Exemplar-based Speech Enhancement. In Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on, pages 2883–2887, Florence, Italy, May 2014.
  12. Deepak Baby and Sibi Raj B. Pillai. Ordered Orthogonal Matching Pursuit. In Communications (NCC), 2012 National Conference on, pages 1–5, Kharagpur, India, Feb 2012.


Tel: +41277217705
Office: 303-6
Contact