Vivek Tyagi, Applied Artificial Intelligence (AI) Research & Development

Education

2006 PhD, School of Computer & Communication Sciences, Swiss Federal Institute of Technology, EPFL, Switzerland

Idiap Research Institute, Eurecom Research Institute

2001 B.Tech./B.S. Electrical Engineering, IIT Kanpur, India

Professional Experience

Please refer to my LinkedIn Page

Awards and Professional Recognition

Co-winner of the International Speech Communication Association (ISCA) Best Paper Award (2009) for the Journal article, ”Automatic speech recognition and speech variability: A review. Speech Communication, Volume 49, Issues 10-11, 2007”

IEEE Senior Member, Signal Processing Society.

Publications:

Recent (Text AI, Speech AI)

1. Vivek Tyagi, Ashvini Jindal, Rajul Anand, Praveen Hegde, Akshat Mathur, "Design and Analysis of High Performance Convolutional Neural Network Text Classifier and its Discriminative Sequence Learning Capability", in ACM Conference on Data Science, Kolkata, Jan 2019, Invited Talk Industrial AI track.

2. Vivek Tyagi, Ashvini Jindal, Rajul Anand, Praveen Hegde, Akshat Mathur, "Design and Analysis of High Performance Convolutional Neural Network Text Classifier and its Discriminative Sequence Learning Capability", in Microsoft Machine Learning and Data Science Conference (MLADS), Redmond, WA, USA, Oct 2018.

• This is a public version of the slides approved for public release by LinkedIn Peer Review with our absolute internal results redacted. I and my team have now successfully deployed 15 CNN Text spam classifiers in English and International (i18n) languages such as French, German, Dutch, Spanish & Portuguese which automatically detect and remove spam content from LinkedIn Feed.

• Business Impact: Based on our CNN Text classification technology, we delivered to our Partner Engineering and Product teams, 30% relative recall gain at 90% precision over the Liblinear and FastText based text classifiers which were the previous production classifiers. In other words, our CNN Text classifiers is able to detect 30% more spam content compared to our previous Liblinear/Fasttext production classifiers at 90% precision.

3. Mohit Yadav, Vivek Tyagi, "Deep Triphone Embedding Improves Phoneme Recognition", arXiv:1710.07868, 2017

4. Vivek Tyagi, "HYBRID CONTEXT DEPENDENT CD-DNN-HMM KEYWORD SPOTTING (KWS) IN SPEECH CONVERSATIONS", To appear in proc. of IEEE Workshop on Machine Learning For Signal Processing (IEEE MLSP 2016).

5. Vivek Tyagi, et al. "Xerox Conversational AI Agent (XCAI) for Enterprise Knowledgebase Q&A", IEEE ICASSP 2016, China, Show and Tell Session

6. V. Tyagi, H. Prasad, T. Faruquie, L. V. Subramaniam, N. Ratha, “Fusing Biographical and Biometric Classifiers for Improved Person Identification”, To appear In the Proc. of IEEE International Conference on Pattern Recognition (ICPR), Nov. 2012, Japan.

7. V. Tyagi and N. Ratha, “Biometric Score Fusion Through Discriminative Training”, In Proc. of IEEE Computer Vision and Pattern Recognition (CVPR) Biometrics Workshop, 2011, Boulder Colorado, USA.

8. Vivek Tyagi, "Fepstrum Features: Design and Application to Conversational Speech Recognition", IBM Research Report No. RI 11009, 6th June 2011

9. B. Srivastava, T. Huan, W. Shang, U. Nambiar, V. Tyagi, S. Kalyanaraman, “Towards a Sustainable Services Ecosystem for Traffic Managament,” In Proc. of Service Research and Innovation Institute (SRII) Global Conference, 2011, San Jose, USA.

10. J. Basak, K. Kate, V. Tyagi, N. Ratha “A Gradient descent approach for multi-modal Biometric Identification”, In the Proc. of IEEE International Conference on Pattern Recognition (ICPR), 2010, Istanbul, Turkey.

11. K. Kate, J. Basak, N. Ratha, V. Tyagi, “QPLC: A novel multimodal biometrics score fusion method”, In the Proc of IEEE Computer Vision and Pattern REcognition (CVPR) Biometrics Workshop, 2010, USA

Journals

1. Vivek Tyagi, Shivkumar Kalyanaraman, Raghu Krishanpuram, " Vehicular Traffic Density State Estimation Based on Cumulative Road Acoustics", In IEEE Transactions on Intelligent Transportation System, 2011.

2. Vivek Tyagi, Herve Bourlard, Christian Wellekens, "On variable-scale piecewise stationary spectral analysis of speech signals for ASR", Speech Communication, Vol. 48 , pages 1182–1191, 2006.

3. Vivek Tyagi, Christian Wellekens, Dirk Slock, "Least squares filtering of speech signals for robust ASR", Speech Communication Vol. 48, pages 1528–1544, 2006

4. M. Benzeghiba, R. De Mori, O. Deroo, S. Dupont *, T. Erbes, D. Jouvet, L. Fissore, P. Laface, A. Mertins, C. Ris, R. Rose, V. Tyagi, C. Wellekens, "Automatic speech recognition and speech variability: A review", Speech Communication Vol. 49, pages 763–786, 2007

Conferences

Speech Recognition

1. 1. V. Tyagi, “Tandem Processing of Fepstrum Features, ” In the Proc. of Interspeech, 2008, Brisbane, Australia.

   2. V. Tyagi, “Maximum Accept and Reject (MARS) training of HMM-GMM speech recognition systems, ”, In the Proc. of Interspeech, 2008, Brisbane, Australia.

   3. V. Tyagi, “Fepstrum: An improved modulation spectrum for ASR, ” In the Proc. of Interspeech, 2007, Antwerp, Belgium.

   4. V. Tyagi and C. Wellekens, “Fepstrum and Carrier Signal decomposition of Speech Signals through Homomorphic Filtering,” In the special session, “Dealing with intrinsic speech variabilities in ASR”, IEEE International Conference on Audio, Speech, and Signal Processing (ICASSP), 2006, Toulouse, France.

   5. V. Tyagi and C.Wellekens, “Fepstrum Representation of Speech,” In the Proc. of IEEE Automatic Speech Recognition and Understanding (ASRU), November 2005, Puerto Rico, USA

   6. V. Tyagi, H. Bourlard and C. Wellekens, “On Variable-Scale Piecewise Stationary Spectral Analysis of Speech Signals for ASR,” In the Proc. of Eurospeech, September 2005, Lisbon, Portugal.

1. 7. V. Tyagi and C. Wellekens, “On Desensitizing the Mel-Cepstrum to Spurious Spectral Components for Robust Speech Recognition,” In the Proc. of IEEE International Conference on Audio, Speech, and Signal Processing (ICASSP), March 2005, Philadelphia, USA.

1. 8. V. Tyagi, I. McCowan, H. Bourlard, H. Misra, “Mel-Cepstrum Modulation Spectrum (MCMS) Features for Robust ASR,” In the Proc. of IEEE Automatic Speech Recognition and Understanding (ASRU), December 2003, St. Thomas, US Virgin Islands.

   9. V. Tyagi, I. McCowan, H. Bourlard, H. Misra, “On Factorizing Spectral Dynamics for Robust Speech Recognition,” In the Proc. of EUROSPEECH, Sept. 2003, Geneva, Switzerland

   10. H. Misra, H. Bourlard, V. Tyagi, “Entropy-Based Multi-Stream Combination,” In the Proc. of IEEE International Conference on Audio, Speech, and Signal Processing (ICASSP), 2003, Hong Kong.

2. Biometrics/Traffic (IBM Smarter Planet Research Theme)

   1. V. Tyagi, H. Prasad, T. Faruquie, L. V. Subramaniam, N. Ratha, “Fusing Biographical and Biometric Classifiers for Improved Person Identification”, To appear In the Proc. of IEEE International Conference on Pattern Recognition (ICPR), Nov. 2012, Japan.

   2. V. Tyagi and N. Ratha, “Biometric Score Fusion Through Discriminative Training”, In Proc. of IEEE Computer Vision and Pattern Recognition (CVPR) Biometrics Workshop, 2011, Boulder Colorado, USA.

   3. B. Srivastava, T. Huan, W. Shang, U. Nambiar, V. Tyagi, S. Kalyanaraman, “Towards a Sustainable Services Ecosystem for Traffic Managament,” In Proc. of Service Research and Innovation Institute (SRII) Global Conference, 2011, San Jose, USA.

   4. J. Basak, K. Kate, V. Tyagi, N. Ratha “A Gradient descent approach for multi-modal Biometric Identification”, In the Proc. of IEEE International Conference on Pattern Recognition (ICPR), 2010, Istanbul, Turkey.

   5. K. Kate, J. Basak, N. Ratha, V. Tyagi, “QPLC: A novel multimodal biometrics score fusion method”, In the Proc of IEEE Computer Vision and Pattern REcognition (CVPR) Biometrics Workshop, 2010, USA