Hardware Aware Training for Efficient Keyword Spotting on General Purpose and Specialized Hardware

Abstract

Keyword spotting (KWS) provides a critical user interface for many mobile and edge applications, including phones, wearables, and cars. As KWS systems are typically 'always on', maximizing both accuracy and power efficiency are central to their utility. In this work we use hardware aware training (HAT) to build new KWS neural networks based on the Legendre Memory Unit (LMU) that achieve state-of-the-art (SotA) accuracy and low parameter counts. This allows the neural network to run efficiently on standard hardware (212$\mu$W). We also characterize the power requirements of custom designed accelerator hardware that achieves SotA power efficiency of 8.79$\mu$W, beating general purpose low power hardware (a microcontroller) by 24x and special purpose ASICs by 16x.

5 pages, TinyML Research Symposium '21

References in corpus (6)

• Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition
• MicroNets: Neural Network Architectures for Deploying TinyML Applications on Commodity Microcontrollers
• Streaming keyword spotting on mobile devices
• TinySpeech: Attention Condensers for Deep Speech Recognition Neural Networks on Edge Devices
• Parallelizing Legendre Memory Unit Training
• Always-On, Sub-300-nW, Event-Driven Spiking Neural Network based on Spike-Driven Clock-Generation and Clock- and Power-Gating for an Ultra-Low-Power Intelligent Device

Cited by in corpus (1)

• Machine Learning for Microcontroller-Class Hardware: A Review