Kannan Achan


Ph.D. candidate

Probabilistic and Statistical Inference Group
University of Toronto

kannan@psi.toronto.edu

My doctoral adviser is Prof. Brendan Frey.  

My research interests are mainly in developing efficient probabilistic inference and machine learning algorithms. Most of my work is geared towards applications in speech/signal processing and computer vision.


Recent Projects
Segmental HMM for Speech Waveforms
Learning to Cluster using Local Neighbourhood Structure
Variational Speech Separation of More Sources than Mixtures
Speech from Spectrogram
Unsupervised Image Translation
Phase Unwrapping
Publications

A Segment-Based Probabilistic Generative Model of Speech
Kannan Achan, Sam Roweis, Aaron Hertzmann, Brendan J. Frey,
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2005). to appear [PS.gz] [PDF]

Variational Speech Separation of More Sources than Mixtures
Steven Rennie, Kannan Achan, Parham Aarabi, Brendan J. Frey,
Artificial Intelligence and Statistics (AISTATS 2005).  [PS.gz | PDF]

Learning to Cluster using Local Neighborhood Structure
Rómer Rosales, Kannan Achan, Brendan J. Frey,
Proc. International Conference on Machine Learning 21 (ICML), Jul 2004.  [PS.gz | PDF]

A Segmental HMM for Speech Waveforms
Kannan Achan, Sam T. Roweis, Brendan J. Frey,
Technical Report UTML-TR-2004-001, University of Toronto, January 2004.  [PS.gz | PDF]

Probabilistic Inference of Speech Signals from Phaseless Spectrograms

Kannan Achan, Sam T. Roweis, Brendan J. Frey,
Neural Information Processing Systems 16 (NIPS 03),
to appear [PS | PDF]

Unsupervised Image Translation

Rómer Rosales, Kannan Achan, and Brendan Frey,  
Proc. International Conference on Computer Vision (ICCV), Nice, France. Oct 2003. 
[PS.gz | PDF

Translating Images by Unsupervised Estimation of Switching Filters
Rómer Rosales, Kannan Achan, and Brendan Frey
(invited) in Proc. IEEE Workshop on Statistical Signal Processing (SSP).  Sep 2003.  
[PS.gz | PDF]

Robust Variational Speech Separation Using  Fewer Microphones Than Speakers
Rennie, S., Aarabi, P., Kristjansson, T., Frey, B., Achan, K.,
Proc. IEEE
International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2003), Hong Kong, April 2003. [PS | PDF]

Phase Unwrapping by Minimizing Kikuchi Free Energy
K. Achan, B. J. Frey, R. Koetter, D. C. Munson, Jr.,
In Proceedings of IEEE IGARSS 2002.

A Factorized Variational Technique for Phase Unwrapping in Markov Random Fields
Kannan Achan, Brendan J. Frey, Ralf Koetter,
In Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence (UAI), Morgan Kaufmann; Seattle, WA, 2001. [PS.gz | PDF]

Unwrapping Phases by Relaxed Mean Field Inference
Kannan Achan, Brendan J. Frey, Ralf Koetter and David C. Munson, Jr.,
International Conference on Acoustics, Speech and Signal Processing, Salt Lake City, UT, 2001. 

Use of Tabu Search in the Logical Topology Design of Wavelength Routed Optical Networks
Kannan Achan, Yash Aneja and Subir Bandyopadhyay, 
In proceedings of the 15th International Symposium on High Performance Computing Systems and Applications, Kluwer Academic Press, 2001

K. Achan, Msc Thesis ``Tabu search in the design of multi-hop optical networks'', April 2000



kannan@psi.toronto.edu
Wednesday, Feb 16th, 2005