Google Inc.

Tianyu Xu

Dongeek Shin

Anuva Kulkarni

Lidan Mu

Jennifer Irturk

Described techniques enable the use of ambisonics encoding/decoding for spatially accurate sound reproduction that is low-latency and that conserves computational and battery resources. Different sounds sources may be encoded using different ambisonics encoding orders, with higher order encoding being used to process sound from a sound source of current attention of a user, and lower order encoding being used to process other sound sources, such as background sound sources.

Saliency-based mixed-order ambisonics encoding of multiple sound sources for immersive spatial audio

Described techniques enable a virtual touchpad for a wearable device by receiving, in response to a touch event, a sensor signal from a motion sensor coupled to the wearable device, the wearable device providing a user interface with a rendering space that is defined relative to the wearable device. The touch event may be associated with a contact space of a contact map defined with respect to the wearable device, based on the sensor signal, and a user interface event of the user interface that corresponds to the contact space may be rendered within the rendering space.

Growing community of inventors

Mountain View, CA, United States of America

Tianyu Xu