Associate Professor, Department of Electrical Engineering
Adjunct Faculty, Samtel Centre for Display Technologies and Flexible Electronics
Email: vipul ar (AT) iitk (.) ac (.) in
Postal Address: 305-D, ACES building, IIT Kanpur - 208016 ()
Office location: Sustainable Energy Engineering Building, Near Media Labs ()
My calendar (office hours: WTh 5-6 PM)
- MADHAV lab: For latest info on my research, team, publications and open positions, please see MADHAV lab.
- DUGC page: I am currently the DUGC convener of EE, IITK.
- MLSP Seminar Series
I work on developing learning based methods mostly for audio processing (music, speech, and other sounds). I have interest in Physics, so I work with Physicists to develop learning based methods for problems in computational Physics too.
I received my B.Tech. and Ph.D. degrees in from the , India. My Ph.D. thesis was titled “”, where I worked on analyzing music audio to identify and transcribe different instruments/voices playing simultaneously. During postdoc at Oxford University (UK), I developed systems using linguistic principles, with applications in automatic language teacher and speech recognition for low-resource languages. At Amazon in Boston (USA), I worked on audio classification for developing Alexa system, with research focusing on classification with imbalanced data.
- Machine learning for audio signal processing
- audio retrieval: multi-label classification
- melody estimation: transfer learning
- speech recognition: confidence estimation
- music analysis: explainable AI
- Machine learning for Physics
- generative modeling, normalizing flows, adversarial learning
- Time series analysis on sensor data
- Sensor calibration: domain adaptation, semi-supervised learning