Google Inc.

Jonathan D Hurwitz

Justin Wooyoung Lee

Diane Wang

Punyabrata Ray

A local and private model can be used to analyze textual, visual, and auditory content on a user's local device to provide notifications about events which are stressful beyond a predetermined threshold. The model can be trained by a combination of health sensor data and an analysis of the content. Multiple models or techniques can be used to analyze the content and vectorize the content. A vector based approach can be used to classify future content as stressful when meeting the predetermined threshold.

Active hidden stressor identification and notification

Embodiments described herein relate to methods, devices, and computer-readable media to determine a compression setting. An input image may be obtained where the input image is associated with a user account. One or more features of the input image may be determined using a feature-detection machine-learning model. A compression setting for the input image may be determined using a user-specific machine-learning model personalized to the user account based on the one or more features in the input image. The input image may be compressed based on the compression setting.

Learning-based image compression setting

The present disclosure relates to methods for monitoring the health of a battery cell. Specifically, the method includes receiving, by a controller communicatively coupled a pressure sensor positioned proximate to a battery housing having an initial volume and an initial shape and containing a battery cell, a pressure value indicative of a current volume or a current shape of the battery' housing. The pressure sensor is configured to measure the pressure value indicative of the current volume or the current shape of the battery' housing. The method also involves comparing, by the controller, the pressure value to a predetermined threshold pressure value. The method further involves adjusting, by the controller and based on the comparison, at least one parameter associated with operations of the battery cell.

Methods to continuously monitor battery cell physical health and swell

An example method includes receiving, by a communication controller of a source device and from an audio source of the source device, audio data to be wirelessly transmitted to a sink device, wherein the audio data is received from the audio source at a rate; storing, by the communication controller and in a buffer, the audio data as a sequence of audio blocks, wherein each audio block includes audio data for a respective output time; transmitting, by the source device and to the sink device, audio blocks of the sequence of audio blocks using a wireless link between the source device and the sink device; determining a current condition of the wireless link; and automatically adjusting, by the communication controller and based on the current condition of the wireless link, the rate at which audio data is received from the audio source.

Radio frequency condition aware audio buffering

An example method includes, responsive to receiving user input to activate an airplane mode of a mobile computing device: disabling a first wireless protocol; determining, based on activity of a media session, whether to maintain an enabled state of a second wireless protocol of the mobile computing device, wherein the mobile computing device is configured to stream audio data associated with the media session to an audio sink device via a second wireless protocol; and selectively maintaining the enabled state of the second wireless protocol based on the determination.

Context aware airplane mode

A connectionless system for handing off data, content or information includes a proximity detection component that allows devices to detect other local devices within range. Devices within range may use advertisement and scanning to exchange communications so that one device can handoff data, content, or information to another device without having to connect, e.g., pair, with the other device(s).

Secure localized connectionless handoffs of data

A method of identifying errors related to a computing device comprising detecting an input to the computing device, comparing the detected input with a threshold, wherein the threshold corresponds to a level of input indicating frustration by a user, determining whether the input meets or exceeds the threshold, and when the input meets or exceeds the threshold, identifying, by the one or more processors, an error related to the computing device.

Growing community of inventors

Cupertino, CA, United States of America

Jonathan D Hurwitz