An Ultra-low Power RNN Classifier for Always-On Voice Wake- Up Detection Robust to Real-World Scenarios

Emmanuel Hardy, Franck Badets

Keywords: Voice Activity Detection (VAD), Wake-Up Sensor, Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), Machine Learning, Internet of Things (IoT), Ultra Low Power, Approximate Quantization, ASIC

TL;DR: We show the superiority of RNNs as ultra-low power classifiers for Voice Wake-Up tasks in terms of detection performance, robustness to unseen noises and low complexity.

Abstract: We present in this paper an ultra-low power (ULP) Recurrent Neural Network (RNN) based classifier for an always-on voice Wake-Up Sensor (WUS) with performances suitable for real-world applications. The purpose of our sensor is to bring down by at least a factor 100 the power consumption in background noise of always-on speech processing algorithms such as Automatic Speech Recognition, Keyword Spotting, Speaker Verification, etc. Unlike the other published approaches, we designed our wake-up sensor to be robust to unseen real-world noises for realistic levels of speech and noise by carefully designing the dataset and the loss function. We also specifically trained it to mark only the speech start rather than adopting a traditional Voice Activity Detection (VAD) approach. We achieve less than 3% No Trigger Rate (NTR) for a duty cycle less than 1% in challenging background noises pooled. We demonstrate the superiority of RNNs on this task compared to the other tested approaches, with an estimated power consumption of 30 nW in 65nm CMOS and a minimal memory footprint of 0.52 kB.