Demo: Active Noise Cancellation on Open-Ear Wearable Devices

Abstract

Active noise cancellation (ANC) is the dominant approach to suppressing environmental noise on modern hearable devices. However, it requires an error microphone at the user's ear canal to measure residual sound. Open-ear wearable devices, such as smart glasses and VR headsets, leave the ear unoccluded and therefore lack ANC functionality. In this demo we present the first ANC system for open-ear wearables that operates using only microphones and open-ear speakers embedded within the frame of the wearables, removing the need for an in-ear error microphone. Our pipeline uses a neural network to estimate the noise at the ear from an array of eight microphones distributed around the wearable's frame and generates an anti-noise signal in real time. The learned microphone-to-ear mapping generalizes to unseen users and acoustic environments without prior acoustic measurement. This demonstration lets users experience our real-time open-ear noise cancellation live. They will hear noise with ANC toggled on and off in various noise conditions and experience how ANC enhances the quality of audio playback through the glasses in a noisy environment.