# Benchmarking Keyword Spotting Efficiency on Neuromorphic Hardware

## TL;DR

This paper benchmarks the efficiency of a neuromorphic chip for keyword spotting, showing it outperforms traditional hardware in energy efficiency while maintaining accuracy, especially with larger networks.

## Contribution

It provides a comparative analysis of neuromorphic hardware versus conventional devices for keyword spotting, highlighting energy efficiency advantages of Loihi.

## Key findings

- Loihi has lower energy cost per inference than CPU, GPU, Jetson TX1, and Movidius.
- Loihi maintains equivalent accuracy to other hardware.
- Energy efficiency advantage increases with larger network sizes.

## Abstract

Using Intel's Loihi neuromorphic research chip and ABR's Nengo Deep Learning toolkit, we analyze the inference speed, dynamic power consumption, and energy cost per inference of a two-layer neural network keyword spotter trained to recognize a single phrase. We perform comparative analyses of this keyword spotter running on more conventional hardware devices including a CPU, a GPU, Nvidia's Jetson TX1, and the Movidius Neural Compute Stick. Our results indicate that for this inference application, Loihi outperforms all of these alternatives on an energy cost per inference basis while maintaining equivalent inference accuracy. Furthermore, an analysis of tradeoffs between network size, inference speed, and energy cost indicates that Loihi's comparative advantage over other low-power computing devices improves for larger networks.