Dr. C. Chandra Sekhar

Professor

Department of Computer Science and Engineering
Indian Institute of Technology Madras
Chennai - 600036,
India.

Office: Room No. 305, Building Sciences Block (BSB)
Phone: +91-44-2257 4363 (Office)
            +91-44-2257 5382 (Speech and Vision Laboratory)
            +91-44-2257 6363 / 2257 0816 (Home)
Fax : +91-44-2257 4352

Email: chandra[AT]cse.iitm.ac.in
URL: http://speech.cse.iitm.ac.in/~chandra

Education:
Research Focus:
Speech Processing, Artificial Neural Networks, Kernel Methods, Pattern Recognition, Content based Information Retrieval, Handwritten Character Recognition

The focus of the research activity is on acoustic modeling of subword units of speech. Approaches using hidden Markov models, multilayer feedforward neural networks and modular neural networks have been explored in developing models for recognition of syllable-like subword units of speech in Indian languages. A constraint satisfaction neural network model that uses constraints based on acoustic-phonetic knowledge to combine evidence from multiple modular networks has been designed for recognition of large number of subword units. Current focus is on acoustic modeling using support vector machines. Issues in developing support vector machine based models for large-class-set pattern recognition and for classification of multidimensional trajectories are being addressed.

Selected Publications:
  1. S. Chandrakala, Veena T. and C. Chandra Sekhar, "Gaussian Mixture Model based Posterior Probability Support Vector Machines for Speech Emotion Recognition and Image Classification", Second International Conference on Intelligent Human Computer Interaction, Allahabad, pp 85-99, January, 2010.

  2. Ved Prakash Sahu, Harendra Kumar Mishra and C.Chandra Sekhar, “Variational Bayes adapted GMM based models for audio clip classification”, International Conference on Pattern Recognition and Machine Intelligence (PReMI’09), New Delhi, pp.513-518, December, 2009.

  3. S.Chandrakala and C.Chandra Sekhar, “Classification of multivariate varying length time series using descriptive statistical features”, International Conference on Pattern Recognition and Machine Intelligence (PReMI’09), New Delhi, pp.13-18, December, 2009.

  4. Dileep A. D. and C. Chandra Sekhar, “Representation and Feature Selection using Multiple Kernel Learning” In proceedings of International Joint Conference on Neural Networks (IJCNN-2009) , Atlanta, USA, pp 717-722, June 2009.

  5. S.Chandrakala and C.Chandra Sekhar, “Combination of generative models and SVM based classifier for speech emotion recognition”, International Joint Conference on Neural Networks (IJCNN-2009), Atlanta, USA, pp.497-502, June, 2009.

  6. Anupam Mandal, C. Chandra Sekhar, K. R. Prasanna Kumar and G. Athithan, "Detection of positions of packet loss in reconstructed VOIP speech", Workshop on Image and Speech Processing, Guwahati, December, 2007.

  7. B. Venkataramana Kini and C. Chandra Sekhar, "Multi-scale kernel latent variable models for nonlinear time series pattern matching", International Conference on Neural Information Processing (LNCS 4985), Kitakyushu, Japan, pp.11-20 (Part II), November, 2007.

  8. Panuku Lakshmi Narayana and C.Chandra Sekhar, "Region-based encoding method using multi-dimensional Gaussians for networks of spiking neurons", International Conference on Neural Information Processing (LNCS 4984), Kitakyushu, Japan, pp.73-82 (Part I), November, 2007.

  9. Anitha Jayaraman, C.Chandra Sekhar and V.Srinivasa Chakravarthy, "Modular approach to recognition of strokes in Telugu script", International Conference on Document Analysis and Recognition, Curitiba, Brazil, pp.501-505, September 2007.

  10. P. Sheetal Reddy, S. Chandrakala and C. Chandra Sekhar, "Local density estimation based clustering", International Joint Conference on Neural Networks, Orlando, Florida, pp.1338-1343, August 2007.

  11. H. Swethalakshmi, Anitha Jayaraman, V. Srinivasa Chakravarthy and C. Chandra Sekhar, "Online handwritten character recognition of Devanagari and Telugu characters using support vector machines", International Workshop on Frontiers in Handwriting Recognition, La Baule, pp.367-372, October, 2006.

  12. Suryakanth V. Gangashetty, C. Chandra Sekhar and B. Yegnanarayana, "Spotting Multilingual Consonant-Vowel Units of Speech Using Neural Networks," Proc. 3rd Int. Conf. Non-linear speech processing, Barcelona, Spain, April 19-22, 2005, pp. 287-297.

  13. D. Srikrishna Satish and C. Chandra Sekhar, "Kernel based clustering and vector quantization for speech recognition," International Workshop on Machine Learning for Signal Processing, Brazil, pp.315-324, September, 2004.

  14. C. Chandra Sekhar, Kazuya Takeda and Fumitada Itakura, "Recognition of subword units of speech using support vector machines," in Recent Research Developments in Electronics and Communication, Transworld Research Network, vol.1, pp.101-136, November, 2002.

  15. C. Chandra Sekhar and B.Yegnanarayana, "A constraint satisfaction model for recognition of Stop Consonant-Vowel (SCV) utterances," IEEE Transactions on Speech and Audio Processing, vol.10, no.7, pp.472-480, October, 2002.

  16. C. Chandra Sekhar, Kazuya Takeda and Fumitada Itakura, "Recognition of Consonant-Vowel (CV) units of speech in a broadcast news corpus using support vector machines," International Workshop on Pattern Recognition using Support Vector Machines, SVM2002, Niagara Falls, pp.171-185, August, 2002.


Teaching Interests:
  1. Pattern Recognition
  2. Artificial Neural Networks
  3. Speech Technology
  4. Computer Architecture
  5. Digital Design

Curriculum Vitae

Links


Last updated: 02-09-2008