$$% Not all commands are supported. Please refer to MathJax docs before including % your macros % http://docs.mathjax.org/en/latest/tex.html#defining-tex-macros % \newcommand{\ALG}{FastGRNN\xspace} \newcommand{\redSpace}{\vspace{-6mm}} \newcommand{\algs}{FastRNN\xspace} \newcommand{\alg}{FastGRNN\xspace} \newcommand{\salg}{FastRNN\xspace} \newcommand{\algfloat}{FastGRNN-LSQ\xspace} %!TEX root = paper.tex \newcommand{\reals}{\mathbb R} \newcommand{\ex}{\mathbb E} \newcommand{\prob}{\mathbb P} \renewcommand{\vec}[1]{{\mathbf{#1}}} \newcommand{\br}[1]{\left({#1}\right)} \DeclareMathOperator{\Tr}{Tr} % Asymptotic notation \newcommand{\bigO}[1]{{\cal O}\br{{#1}}} \newcommand{\softO}[1]{\widetilde{\cal O}\br{{#1}}} \newcommand{\Om}[1]{\Omega\br{{#1}}} \newcommand{\softOm}[1]{\tilde\Omega\br{{#1}}} \def\half{{\textstyle\frac{1}{2}}} \newcommand{\smallfrac}[2]{{\textstyle \frac{#1}{#2}}} \def\v#1{\vec #1}$$

# Projects

GesturePod is a plug-and-play, gesture recognition device that is designed to be clamped onto any white-cane used by persons with Visually Impairment. Once clamped onto the cane firmly, simple and natural gestures performed on the cane can be used to interact with various devices, for instance a mobile phone.
EdgeML enables small, fast and accurate classifiers based on LSTM and ProtoNN for real-time keyword spotting on Raspberry Pi3 and Pi0. Our latest set of works, (EMI-RNN and Shallow RNNs) makes keyword spotting possible on even smaller devices; as small as a MXChip with a Cortex M4.