Anurag Katakkar
I am a Deep Learning Engineer at Matroid. Previously, I was a Deep Learning Software Engineer at NVIDIA, where I worked on the TensorRT Team.
Before that, I graduated from Carnegie Mellon University with a Master of Science in Computational Data Science. I was advised by Alan W Black, and Eric Nyberg.
My primary research area is Speech Processing. Specifically, I'm interested in modeling speech directly from waveforms/spectrograms. Towards that goal, I'm interested in exploring self-supervised learning based approaches. I also care about the applications of speech technology to social good and recently organized the Speech for Social Good Workshop (S4SG) at Interspeech 2022.
I'm also broadly interested in building efficient Machine Learning Systems - through algorthmic breakthroughs and better software for machine learning.
News
My paper was accepted at the Blackbox NLP workshop at EMNLP 2022! (Practical Benefits of Feature Feedback Under Distribution Shift, pdf)
Started working as a Deep Learning Engineer at Matroid!
Organized the Speech for Social Good Workshop at Interspeech 2022
Towards Language Modelling in the Speech Domain Using Sub-word Linguistic Units is now available on arXiv.
My paper Practical Benefits of Feature Feedback Under Distribution Shift is now available on arXiv.
I've started working as a Deep Learning Software Engineer, TensorRT, at NVIDIA!
Graduated from CMU with a Master's in Computational Data Science!