Sound Sensing System Using Machine Learning Accelerator

Abstract

Machine learning allows for us to interact with data in a meaningful way on a very large scale. A the amount data available to us grows, so does the potential of machine learning. However, there is a drawback in that the computations needed to draw conclusions from this data can be extremely expensive. This paper explores a potential use of a machine learning accelerator that allows us to do these computations in a more efficient manner.