Research

My motto: Is to design and apply technologies to ease livability

I started my research career in August 2010 when I joined IISc Bangalore. My research interests include Machine Listening, Audio Source Separation, Speech Signal Processing, Audio and Speech Analysis, Machine Learning, Deep Learning, Data Analytics, Image Processing and Sparse Signal Processing

Links

Indic keyboards on Windows and Linux (from MILE Lab IISc) - https://code.google.com/p/indic-keyboards/

Kannada TTS web demo: http://mile.ee.iisc.ernet.in:8080/tts_demo/

MILE Indic keyboards: https://play.google.com/store/apps/details?id=org.iisc.mile.indickeyboards.android

Cool Audio Demos: http://www.cs.illinois.edu/~paris/demos/

Blind Source Separation: http://www.ism.ac.jp/~shiro/research/blindsep.html, http://web.media.mit.edu/~paris/ica.html

Nice and creative depiction of speech signals in one and two dimensions: http://www.youtube.com/watch?v=YTw8aMiR83M&feature=youtu.be

My research interests include Machine Listening, Audio Source Separation, Speech Signal Processing, Audio and Speech Analysis, Machine Learning, Deep Learning, Data Analytics, Image Processing and Sparse Signal Processing

I graduated from National Institute of Technology Karnataka , Surathkal in 2008 with a B.Tech in EEE. I have gained knowledge and experience on various topics spanning electrical, electronics and computer science domains during my 4 years of B.Tech. I really liked the flexibility and variety of courses and projects done during my B.Tech which gave me a wide foundation to pursue higher studies.

Projects during B.Tech August 2004 - May 2008:

• Major Project: Steady State Analysis of Unified Power Flow Controller using Matlab 5.3 and Simulink

3.0- group project under Dr. K.N. Shubhanga of Dept of Electrical and Electronics Engineering, NITK

Surathkal

• Digital Quiz- group project under Mr. K. Manjunath Sharma of Dept of Electrical and Electronics

Engineering, NITK Surathkal

• Pspice Simulation of high frequency signal only during negative half cycle of the power supply- individual

project under Mr. A.R.Beig of Dept of Electrical and Electronics Engineering, NITK Surathkal

• Design and implementation of Badminton Scoreboard (for doubles match) using Xilinx VHDL softwareindividual

project under Prof. P.Vittal of Dept of Electrical and Electronics Engineering, NITK Surathkal

• Graphic Equalizer implementation using Matlab 7.0 - individual project under Prof. Jora M Gonda of

Dept of Electrical and Electronics Engineering, NITK Surathkal

• Design of a a simplified bus model of Embedded Intel486 SX Processor using Xilinx VHDL softwaregroup

project under Prof. P.Vittal of Dept of Electrical and Electronics Engineering, NITK Surathkal

• 2-Dimensional electromagnetic field simulation for high performance electromechanical design using

Maxwell SV software- group project under Dr.P.Duraikannu of Dept of Electrical and Electronics

Engineering, NITK Surathkal

• Speed Control of DC Motor using PC Generated Pulse Width Modulation (Hardware Project)- group

project under Mrs. Vinatha U. of Dept of Electrical and Electronics Engineering, NITK Surathkal

• Implementation of Image Processing Techniques using Graphical User Interface in Matlab 7.0.1- group

project under Mrs. Vinatha U. of Dept of Electrical and Electronics Engineering, NITK Surathkal

I worked in L&T, Mumbai for around 2 years before joining Dept of Electrical Engineering,Indian Institute of Science for my MSc(Engg) in 2010. I upgraded my registration to PhD in March 2012. During my PhD, I took various courses and gained teaching and mentoring experience. I explored and proposed various applications in speech and noise analysis using signal processing and machine learning techniques. My PhD thesis is titled as "Speech and noise analysis using sparse representation and acoustic phonetics knowledge". I submitted my thesis on 17 April, 2017 and defended on 25 September, 2017. A copy of my PhD thesis is here.

Projects during PhD August 2010- April 2017:

• Relationship Between Indian Languages Using Bigram Language Models- Group project under Prof. T.

V. Sreenivas of Dept of Electrical Communication Engineering, Indian Institute of Science

• Voiced and Unvoiced feature classification of Speech data using multiple features- Individual course

project under Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

• Implementation of Modification of Pitch contour , Duration and Energy for change in Prosody in Tamil

Speech Synthesis under Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of

Science

• Voice Modification system using change of sampling frequency and duration normalization under Prof.

A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

• Implementation of Parts of Speech and Pause Rules Tagging of Tamil text in Matlab and C under Prof.

A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

• Robust estimation of glottal closure instants using dynamic weighting of subband components in

Speech under Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

• Subband Analysis of Linear Prediction Residual for the Estimation of Glottal Closure Instants under Dr.

T. V. Ananthapadmanabha of Voice and Speech Systems, Bangalore and Prof. A. G. Ramakrishnan of

Dept of Electrical Engineering, Indian Institute of Science

• Detection of transitions among broad phonetic classes in a speech signal using temporal features and

a rule based approach under Dr. T. V. Ananthapadmanabha of Voice and Speech Systems, Bangalore

and Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

• A dictionary learning and source recovery based approach to classify diverse audio sources under Dr.

T. V. Ananthapadmanabha of Voice and Speech Systems, Bangalore and Prof. A. G. Ramakrishnan of

Dept of Electrical Engineering, Indian Institute of Science

• Hierarchical classification of speaker and background noise and estimation of SNR using sparse representation

under Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian Institute of Science

and Dr. T. V. Ananthapadmanabha of Voice and Speech Systems, Bangalore

• Supervised and adaptive dictionary learning using sparse representation for noisy speech analysis

and language classification under Prof. A. G. Ramakrishnan of Dept of Electrical Engineering, Indian

Institute of Science and Dr. T. V. Ananthapadmanabha of Voice and Speech Systems, Bangalore

• Relationship between spoken Indian languages by clustering of long distance bigram features of speech,

I have worked in Huawei Technologies India as a Senior Technical Leader on machine learning for various cloud computing applications and developed a face recognition system for IOT.

Resume

Linkedin profile