Raise to Speak
Shiwen Zhao, Brandt Westing, Shawn R. Scully, Heriberto Nieto, Roman Holenstein, Minwoo Jeong, K. P. Sridhar, Brandon Newendorp, Mike Bastian, Sethu Raman, Tim Paek, Kevin R. Lynch, Carlos Guestrin · 2019
The two most common ways to activate intelligent voice assistants (IVAs) are button presses and trigger phrases. This paper describes a new way to invoke IVAs on smartwatches: simply raise your hand and speak naturally. To achieve this experience, we designed an accurate, low-power detector that works on a wide range of environments and activity scenarios with minimal impact to battery life, memory footprint, and processor utilization. The raise to speak (RTS) detector consists of four main compo- nents: an on-device gesture convolutional neural network (CNN) that uses accelerometer data to detect specific poses; an on-device speech CNN to detect proximal human speech; a policy model to combine signals from the motion and speech detector; and an off-device false trigger mitigation (FTM) system to reduce unin- tentional invocations trigged by the on-device detector. Majority of the components of the detector run on-device to preserve user privacy. The RTS detector was released in watchOS 5.0 and is running on millions of devices worldwide.