University of California

Li Zhang

Prasant Mohapatra

Faryar Jabbari

Parth H Pathak

Scott Samuelsen

Muchen Wu

The disclosed embodiments relate to the design of a system that uses an accelerometer in a mobile device to detect hotwords, which activate a voice interface to recognize subsequent voice input. During operation, the system gathers samples comprising readings from the accelerometer in the mobile device. Next, the system calculates features from the gathered samples. The system then classifies the calculated features using a classifier, which has been trained to detect hotwords based on calculated features. Finally, if the classifier detects one or more hotwords, the system causes the mobile device to launch a voice-control system that recognizes subsequent voice input received from an audio microphone.

Energy-efficient, accelerometer-based hotword detection to launch a voice-control system

A method of optimizing electricity use from an electric grid is disclosed. The method includes formulating a target load using a cost function for the electric grid, broadcasting a cost signal to the electric device based on the target load, receiving a charging profile from the electric device based on the cost signal, aggregating a demand profile for the electricity use based on the charging profile, determining an updated cost signal based on the demand profile, and broadcasting the updated cost signal to another electric device. The electric device is chargeable electric device, such as a plug-in electric vehicle (PEV). The electric device determines the charging profile only once based on the cost signal, and is charged at maximum power.

Growing community of inventors

Menlo Park, CA, United States of America

Li Zhang