AFSANEH ASAEI

Education

Honors

Academic Service

Publications

Professional Experiences

Contact

Hi Dear, welcome to my homepage! I am a postdoctoral researcher at Idiap Research Institute, with supervision of Prof. Herve BourlardThe subject of my research is parsimonious and deep learning for robust speech processing and recognition. The past fourteen years of my professional work have been devoted to various disciplines in signal processing and machine learning with a focus on deep learning, sparse component analysis, microphone arrays and statistical pattern recognition. Currently, my passion is identifying a domain where machine listening paradigms can be realized for adverse (e.g. multiparty) conditions. My research interests lie in the broad area of signal processing, machine learning, statistics, acoustics, auditory scene analysis and cognition, sparse signal recovery and acquisition. CV

 

 

Education

Thesis: “Model-based Sparse Component Analysis for Multi-party Distant Speech Recognition

Thesis: “Sound Localization by Beamforming Techniques for Robust Speech Recognition

Thesis: “Design and Implementation of a Band-Pass FIR Filter using TMS320C57 DSP processor

 

Professional Experience

  • Idiap Research Institute, Martigny, Switzerland, 2008-Current

     

    Postdoctoral Researcher at Speech & Audio Group

    Areas of Research & Supervision:

      o   Sparse and Low-rank Modeling of DNN posteriors for Robust Speech Recognition

      o   Subspace Modeling of DNN posteriors for Keyword Detection

      o  Exploiting Structured Sparsity of DNN Phonological Representations for Speech Coding

     

    Research Assistant at Speech & Audio Group

    Areas of Research:

      o   Acoustic Informed Source Localization and Separation

      o   Sparse Component Analysis exploiting Auditory and Image (reverberation) Models

      o  Multiparty Distant Speech Recognition

     

  • Iran Telecommunication Research Centre (ITRC), Tehran, Iran, 2002-2008

     

    Multimedia Systems Research Group

     

            Member of the “Biometrics” research team, 2007-2008

            Initiated our team project on multimodal biometrics for web-based applications

            Areas of Research:

            o   Speaker Verification

            o   Classifier fusion

     

                     Member of the “Electronic Health Research Management” team, 2005-2007

            Assessment of the outsourced projects and proposals on:

            o   Blind Source Separation

            o   Source Localization and Tracking

            o   Automatic Speech Recognition

     

    Voice Over IP Project

     

           Supervisor of the “DSP” team in the following projects, 2003-2004

                 o   Design and Simulation of a Fixed-Point G.168 Multichannel Echo Canceller for Voice over IP Media Gateway on TMS320C6416 EVM platform

            o   Design and Simulation of a Fixed-Point Fax Tone Detector and Generator on TMS320C6416 code composer simulator

     

          Supervisor of the “PCI Device Driver Writing” team, 2003

            o  Writing a PCI Device Driver on Linux for TMS320C6416 EVM platform.

            o  PCI Device Driver on Windows Platforms

     

       o   Implementation of a TMS320C542 laboratory board communicating with PC through the DSP HPI port, 2002

       o   Implementation of a voice loop back  using codec & TMS320C542 DSP processor, 2001

           

  TOP

Honors

 

       

    TOP

Publications 

Patents

      Method, apparatus and computer program product for determining the location of a plurality of speech sources, 2012US-13/654055, October 2012

      Signal processing method and apparatus based on structured sparsity of phonological features, US2015846036 (14/846,036), 04.09.2015.

Journal Papers

      “Sparse Modeling of Neural Network Posterior Probabilities for Exemplar-based Speech Recognition”, with P. Dighe, H. Bourlard, Speech Communication, vol 76, 230-244, 2016.

      “Structured Sparsity Models for Reverberant Speech Separation”, with M. Golbabaee, H. Bourlard, V. Cevher, IEEE/ACM Transactions on Speech and Audio Processing, , 22 (3), pp 620-633, 2014. [bibtex]

      “Convexity in Source Separation: Models, geometry and algorithms”, with M. B. McCoy, V. Cevher, Q. T. Dinh, L. Baldassarre, IEEE Signal Processing Magazine, Special Issue on Source Separation and Applications, 31 (3), 87-95, 2014.

      “Ad Hoc Microphone Array Calibration: Euclidean Distance Matrix Completion Algorithm and Theoretical Guarantees”, with M. J. Taghizadeh, R. Parhizkar, P. N. Garner and H. Bourlard, Signal Processing, vol 107, 123-140, 2014.

      “Computational Methods for Underdetermined Convolutive Speech Localization and Separation via Model-based Sparse Component Analysis”, with H. Bourlard, M. Taghizadeh, V. Cevher, Speech Communication, vol 76, 201-217, 2016.

      “Verified Speaker Localization Utilizing Voicing Level in Split-bands”, with M. J. Taghizadeh, M. Bahrololum, M. Ghanbari,  Signal Processing,  vol. 89, Issue 6, pp 1038-1049, June 2009. [bibtex]

      “Binary Sparse Coding of Convolutive Mixtures for Sound Localization and Separationvia Spatialization”, with H. Bourlard, B. Raj, M. Taghizadeh, V. Cevher, IEEE Transactions on Signal Processing, 64 (3), 567-579, 2016.

      “Spatial Sound Localization via Multipath Euclidean Distance Matrix Recovery”, with M. J. Taghizadeh, S. Haghighatshoar, P. N. Garner and H. Bourlard, IEEE Journal of Selected Topics in Signal Processing, vol. 9, 802-814, 2015.

      “TDOA Matrices: Algebraic Properties and their Application to Robust Denoising with Missing Data”, with J. Velasco, D. Pizarro, J. Macias-Guarasa, IEEE Transactions on Signal Processing,, Issue 99, July 2016.

      “Composition of Deep and Spiking Neural Networks for Very Low Bit Rate Speech Coding”, with M. Cernak, A. Lazaridis, P. N. Garner IEEE/ACM Transactions on Speech and Audio Processing, (to appear) 2016.

      “On Structured Sparsity of Phonological Posteriors for Linguistic Parsing”, with M. Cernak, H. Bourlard, Speech Communication, (to appear) 2016.

Conference and Workshop Papers:

          “Low-Rank Representation of Nearest Neighbor Phone Posterior Probabilities to Enhance DNN Acoustic Modeling”, with G. Luyet, P. Dighe, H. Bourlard,  INTERSPEECH, 2016.

      “Phonetic and Phonological Posterior Search Space Hashing Exploiting Class-Specific Sparsity Structures”, with G. Luyet, M. Cernak, H. Bourlard,  INTERSPEECH, 2016.

      “Sound Pattern Matching for Automatic Prosodic Event Detection”, with M. Cernak, P. Honnet, P. N. Garner, H. Bourlard,  INTERSPEECH, 2016.

      “Subspace Detection of DNN Posterior Probabilities via Sparse Representation for Query by Example Spoken Term Detection”, with D. Ram, H. Bourlard,  INTERSPEECH, 2016.

      “Exploiting Low-dimensional Structures to Enhance DNN based Acoustic Modeling in Speech Recognition”, with P. Dighe, G. Luyet, H. Bourlard,  ICASSP, 2016.

      “On Compressibility of Neural Network phonological Features for Low Bit Rate Speech Coding”, with M. Cernak, H. Bourlard,  INTERSPEECH, 2015.

      “Sparse Modeling of Posterior Exemplars for Keyword Detection”, with D. Ram, H. Bourlard,  INTERSPEECH, 2015.

      “Dictionary Learning for Sparse Representation of Neural Network Exemplars in Speech Recognition”, with P. Dighe, H. Bourlard,  Signal Processing with Adaptive Sparse Structured Representations (SPARS), 2015.

      “Sparse Modeling of Neural Network Posterior Probabilities for Exemplar-Based Speech Recognition”, with P. Dighe, H. Bourlard,  Signal Processing with Adaptive Sparse Structured Representations (SPARS), 2015.

      “On Application Of Non-Negative Matrix Factorization for Ad Hoc Microphone Array Calibration from Incomplete Noisy Distances”, with N. Mohammadiha, M. J. Taghizadeh, S. Doclo, H. Bourlard,  ICASSP, 2014.

      “Robust Microphone Placement for Source Localization from Noisy Distance Measurements”, with M. J. Taghizadeh, S. Haghighatshoar, P. N. Garner and H. Bourlard,  ICASSP, 2014.

      “Novel GCC-PHAT Model in Diffuse Sound Field for Microphone Array Pairwise Distance Based Calibration”, with J. Velasco, M. J. Taghizadeh, H. Bourlard, C. J. Martin-Arguedas, J. Macias-Guarasa, D. Pizarro,  ICASSP, 2014.

      “Posterior-based Sparse Representation for Automatic Speech Recognition ”, with S. Bahaadini, D. Imseng and H. Bourlard,  Intl. Speech Communication Association INTERSPEECH, Singapore, September, 2014.

      “Ad-Hoc Microphone Array Calibration from Partial Distance Measurements”, with M. J. Taghizadeh, P. N. Garner and H. Bourlard,  Proceedings of the 4th Joint Workshop on Hands-free speech communication and Microphone Arrays (HSCMA), Nancy, France, May, 2014; nominated for best student paper award

      “Model-based Sparse Component Analysis for Reverberant Speech Localization”, with H. Bourlard, M. J. Taghizadeh, V. Cevher,  Intl. Conference on Acoustic Speech and Signal Processing (ICASSP), Florence, Italy, May, 2014.

     Manifold Sparse Beamforming”, with B. Gozcu and V. Cevher, IEEE International Workshop on Computational Advances in Multi-sensor Adaptive Processing, Saint Martin, May 12-14, 2013; invited paper .       

      “Structured Sparse Acoustic Modeling for Speech Separation”, with M. Golbabaee, H. Bourlard, V. Cevher, Signal Processing with Adaptive Sparse Structured Representations (SPARS), Lausanne, Switzerland, July, 2013. [bibtex]

      A Multipath Sparse Beamforming Method”, with B. Raj, H. Bourlard, V. Cevher, Signal Processing with Adaptive Sparse Structured Representations (SPARS), Lausanne, Switzerland, July, 2013. [bibtex]

      “Structured Sparse Coding for Microphone Array Location Calibration”, with B. Raj, H. Bourlard, V. Cevher, SAPA-SCALE Conference, Intl. Speech Communication Association, Portland, OR, September, 2012. [bibtex]

      Computational Methods for Structured Sparse Component Analysis of Convolutive Speech Mixtures”, with M. E. Davies, H. Bourlard, V. Cevher,  Intl. Conference on Acoustic Speech and Signal Processing (ICASSP), Kyoto, Japan, March, 2012. [bibtex]

      Multi-party Speech Recovery Exploiting Structured Sparsity Models”, with M. J. Taghizadeh, H. Bourlard, V. Cevher, Intl. Speech Communication Association, INTERSPEECH, 2011. [bibtex] - Demo

      An Integrated Framework for Multi-Channel Multi-Source Localization and Voice Activity Detection”, with M. J. Taghizadeh, P. N. Garner, H. Bourlard, H. R. Abutalebi, Third Joint Workshop on Hands-free Speech Communication and Microphone Arrays (HSCMA), Edinburgh, Scotland, 2011. [bibtex]

      “Model-Based Compressive Sensing for Multi-Party Distant Speech Recognition”, with H. Bourlard, V. Cevher, Intl. Conference on Acoustic Speech and Signal Processing (ICASSP), Prague, Czech Republic, May 2011; winner of IEEE spoken language processing award. [bibtex] - Demo

            Sparse Component Analysis for Speech Recognition in Multi-speaker Environment”, with H. Bourlard and P. N. Garner, Intl. Speech Communication Association, INTERSPEECH, 2010. [bibtex]

      Analysis of Phone Posterior Feature Space Exploiting Class-Specific Sparsity and MLP-Based Similarity”, with B. Picart, H. Bourlard, , Intl. Conference on Acoustic Speech and Signal Processing (ICASSP), Dallas, Texas, March 2010. [bibtex]

      Far-field Continuous Speech Recognition System based on Speaker Localization and Sub-band Beamforming”, with M. J. Taghizadeh, H. Sameti, , Proceedings of The 6th International ACS/IEEE Conference on Computer Systems and Applications ,Doha, Qatar, pp 495-500, April 2008. [bibtex]

      “Speaker Direction Finding for Practical Systems: A Comparison of Different Approaches”, with S. Ghanbari, M. J. Taghizadeh, H. Sameti, , Proceeding of the third Annual IEEE BENELUX/DSP valley signal processing symposium, Metropolis, Antwerp, Belgium, pp 129-133,  March 2007. [bibtex]

      Robust Speaker Localization Utilizing a Novel Beamforming Algorithm Based on Harmonic Structures”, with H. Sameti, M. Sh. Moin, , Proceeding of the 12th International Computer Society of Iran Computer Conference (CSICC'07), Tehran, Iran, February 2007. [bibtex]

Workshop

      “Speaker Verification and Localization by Microphone Array”, Presented in a one-day workshop on Biometrics at the 15th Iranian Conference on Electrical Engineering (ICEE’07), Tehran, May 12th, 2007

Technical Report

      Sparse Hidden Markov Models for Exemplar-based Speech Recognition Using Deep Neural Network Posterior Features”, with P. Dighe, H. Bourlard, EPFL-REPORT-210404, 2015

      A New Identity for the Least-square Solution of Overdetermined Set of Linear Equations”, with S. Haghighatshoar, M. J. Taghizadeh, arXiv:1502.07695, 2015

      Structured Sparse Component Analysis of Compressive Acoustic Measurements”, with H. Bourlard and V. Cevher, IDIAP-RR-Internal, 2011, September 2011

       Investigation of kNN Classifiers on Posterior Features Towards Application in Automatic Speech Recognition, with H. Bourlard and B. Picart, IDIAP-RR-76, May 2009

       Kalman and Extended Kalman Filters for Source Tracking, ITRC Technical report, Multimedia Systems Research Group, August 2006

       Speech Recognition Systems: A Survey Study, ITRC Technical report, Multimedia Systems Research Group,  May 2006

      Voice Activity Detection: The Practical Solution for a Continuous Speech Recognition Engine, ITRC Technical report, Multimedia Systems Research Group,  September 2005

       Test Procedure for G.168 and G.729 ITU-T Recommendations, ITRC Technical report, VOIP Project, December 2004

       Design and Simulation of a Fixed-Point G.168 Multichannel Echo Canceller for TMS320C6416 EVM platform, ITRC Technical report, VOIP Project,  October 2004

        Fax Tone Detection and Generation, Design and Simulation on the TMS320C6416 code composer, ITRC Technical report, VOIP Project, December 2003

       How to Write a Linux PCI Device Driver for TMS320C6416 EVM platform, ITRC Technical report, VOIP Project, May, 2003

       2040 PCI Controller for DSP Boards, ITRC Technical report, VOIP Project, January 2003

 

  TOP

Academic Service

Referee

      IEEE Signal Processing Magazine

      Neural Information Processing Systems - NIPS 2016

      IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP 2016

      Speech Communication

      IET Communications

      Frontiers of Information Technology & Electronic Engineering

      Multidimensional Systems and Signal Processing Journal

      IEEE Sensor Array and Multichannel Signal Processing Workshop - SAM 2016

      IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing - CAMSAP 2015 (TPC)

      The 12th International Conference on Latent Variable Analysis and Signal Separation - LVA/ICA 2015 (TPC)

      The Fourth Joint Workshop on Hands-Free Speech Communication and Microphone Arrays - HSCMA 2014 (TPC)

      IEEE Transactions on Aerospace and Electronic Systems

      Research Foundation - Flanders (Fonds Wetenschappelijk Onderzoek - Vlaanderen), FWO

      European Signal Processing Conference (EUSIPCO 2015-2016)

      The Eighth IEEE Sensor Array and Multichannel Signal Processing Workshop

      IEEE International Symposium on Information Theory

      IEEE Journal of Selected Topics in Signal Processing

      ACM transactions of speech and language processing

      Neural Computing and Applications Journal

      IEEE Transactions on Signal Processing

      Multidisciplinary Digital Publishing Institute (MDPI), Sensors

      EURASIP Journal on Advances in Signal Processing

      IEEE Signal Processing Letters

      IEEE Transactions on Audio, Speech and Language Processing

      IEEE International Workshop on Computational Advances in Multi-sensor Adaptive Processing

      IEEE Workshop on Applications of Signal Processing to Audio and Acoustics

      International Conference on Multimedia Computing and Information Technology

      2012 IEEE International Symposium on Circuits and Systems

Meetings and presentations

      Summer School on Machine Learning in Cambridge, UK, 29 August to 10 September 2009

      SCALE winter school on distant speech recognition at Saarland university, Germany, 11-22 January, 2010

      ICASSP conference in Dallas, Texas, USA, 14-19 March, 2010

      Google Scholars Retreat at Google office in Zurich, Switzerland, 28-30 June, 2010

      INTERSPEECH Conference in Makuhari, Japan, 26-30 September, 2010

      HSCMA workshop, Edinburgh, Scotland, 30 May to 1 June, 2011

      SPARS'11 workshop, Edinburgh, Scotland, 27-30 June, 2011

     INTERSPEECH Conference, Florence, Italy, 28-31 August, 2011, photos

      Interactive Multimodal Information Management Summer Institute, Switzerland, 2010-2011, photos

      A Compressive Sensing Perspective on Random Sensor Array (Edinburgh - Sparsity reading group) slides

      SCALE winter school on "Beyond HMM" at Radboud university, Nijmegen, 24-27 January, 2012

          -  Session Chair: Keynote on "Exemplar-based methods for automatic speech recognition"

      SPARS'13 workshop, Lausanne, Switzerland, 8-11 July, 2013

      DSP'13 IEEE International Conference on Digital Signal Processing, Santorini, Greece, 1-3 July, 2013

 

      Sharon Gannot, Afsaneh Asaei, LVA/ICA'15 Special Session on Sparse modeling and low-rank modeling for acoustic speech processing, Liberec, Czech Republic, August 25-28, 2015.

      Herve Bourlard, Tara Sainath, Sharon Gannot, Afsaneh Asaei, Guest Editor of the ELSEVIER Journal of Speech Communication, Special Issue on Advances in Sparse Modeling and Low-rank Modeling for Speech Processing, August, 2014.

      Herve Bourlard, Afsaneh Asaei, HSCMA'14 Special Session on Advances in sparse modeling and low-rank modeling for speech processing, Nancy, France, May 12-14, 2014 CFP

      IBM Watson Research Center, Yorktown Heights, NY, USA, 13 June, 2012 - Invited Visit

          -  Presentation: Model-based Sparse Component Analysis for Recovering Multi-Party Speech from Multi-channel Recordings

      Machine Learning Workshop, EPFL, Switzerland, 19 November, 2012  - Invited Talk

        -  Presentation: Structured Sparse Coding for Machine Listening

 

  TOP

Contact

 

  TOP

This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All person copying this information are expected to adhere to the terms and constraints invoked by each document's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

Last update on April 11, 2016