Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech synthesis. The methods, systems, and apparatus include actions of obtaining an audio representation of speech of a target speaker, obtaining input text for which speech is to be synthesized in a voice of the target speaker, generating a speaker vector by providing the audio representation to a speaker encoder engine that is trained to distinguish speakers from one another, generating an audio representation of the input text spoken in the voice of the target speaker by providing the input text and the speaker vector to a spectrogram generation engine that is trained using voices of reference speakers to generate audio representations, and providing the audio representation of the input text spoken in the voice of the target speaker for output.

Abstract: A control system includes an energy management system in operation with intelligent, network-connected thermostats located in structures. The thermostats are operable to control heating, ventilation, and air conditioning (HVAC) systems. Control during a demand response (DR) event period may be performed based on an optimal setpoint schedule of the HVAC system. Candidate setpoint schedules may be simulated to identify an optimal schedule.

Abstract: Systems, methods, and media for presenting media content are provided. In some implementations, a system for presenting media content is provided, the system comprising: a hardware processor that is programmed to: establish a connection with a media playback device; respond to a discovery request from a computing device presenting a media content item; establish a persistent communication channel with the computing device over a network in response to a request from the computing device to establish the persistent communication channel; receive identifying information of the media content item being presented by the computing device and identifying information of a source of the media content item being presented by the computing device; request the media content item from the identified source of the media content item as a stream of media content; and cause the media content item to be presented using the media playback device.

Abstract: Techniques for performing an emissions demand response event are described. In an example, a power control server system receives an emissions rate forecast for a predefined future time period. Using the emissions rate forecast, an emissions rate event is identified during the predefined future time period. Based on the plurality of emissions rate event, an emissions demand response event is generated during the predefined future time period. The power control server system then causes a power controller to modify an energy consumption by an electronic device in accordance with the generated emissions demand response event.

Abstract: Implementations of the disclosure provide for anticipatory pre-execution of queries. In an implementation of the disclosure, a method for anticipatory pre-execution of queries includes the computation of an execution cost of each of a multiplicity of different queries to a data source scheduled at a specified time on a specified date. The method also includes monitoring a querying processing schedule and detecting from the monitoring, unscheduled time on a particular date prior to the specified time on the specified date. Finally, the method includes responding to the detection by first selecting a most expensive one of the different queries in terms of execution cost, for instance an amount of computing resources consumed in executing a corresponding one of the different queries, and then executing the selected most expensive one of the different queries during the unscheduled time on the particular date prior to the specified time on the specified date.

Abstract: A personal electronic device is provided with a feature of intelligently adjusting a sampling rate of sensors, such as barometers. Before acquiring a measurement from the sensor, the device checks other hardware for conditions that could introduce noise into the measurement. If the device does not determine that excessive noise is likely after checking the other hardware, the device proceeds to acquire the measurement from the sensor. Otherwise, the device delays the measurement. At the end of the delay, the device may check the other hardware again before acquiring the measurement from the sensor.

Abstract: Embodiments are provided for managing the operation of sensors in an electronic device. According to certain aspects, the electronic device may detect a change in motion from a set of lower-sensitivity sensor data generated by a sensor(s) operating in a lower-sensitivity mode. When the change in motion is detected and during a timeout window, the sensor(s) may generate an additional set of lower-sensitivity sensor data and a set of higher-sensitivity sensor data. The electronic device may initially confirm the change in motion based on analyzing the set of higher-sensitivity sensor data. Further, the electronic device may determine that the additional set of lower-sensitivity does not indicate an additional change in motion, and may deem the confirmation of the change in motion as a false positive.

Abstract: This disclosure relates to a privacy preserving machine learning platform. In one aspect, a method includes receiving, from a client device and by a computing system of multiple multi-party computation (MPC) systems, a first request for user group identifiers that identify user groups to which to add a user. The first request includes a model identifier for a centroid model, first user profile data for a user profile of the user, and a threshold distance. For each user group in a set of user groups corresponding to the model identifier, a centroid for the user group that is determined using a centroid model corresponding to the model identifier is identified. The computing system determines a user group result based at least on the first user profile data, the centroids, and the threshold distance. The user group result is indicative of user group(s) to which to add the user.

Abstract: A system and method for interfacing a television with a second device is provided. The system includes an interfacing unit to enter an interfacing mode between the television mode and the second device; a card storage unit to store a first card and a second card, the first card being presently displayed on the television and the second card being previously displayed on the television; a current card display register to store the first card; an interactive retrieval unit to retrieve a plurality of interactive functions associated with the first card; and a display unit to display the first card and the plurality of interactive functions in response to being in the interfacing mode.

Abstract: This document describes techniques and apparatuses that include a three-dimensional (3D) antenna module for transmitting or receiving electromagnetic millimeter waves (mmWaves). In general, a user equipment (UE) may include the 3D antenna module in a corner of a housing of the UE. The 3D antenna module may include three antenna panels that are generally planar and generally orthogonal to three respective axes of a Cartesian-coordinate system. The 3D antenna module may transmit and receive the electromagnetic mmWaves as part of a wireless link between the UE and another device, such as a satellite that is part of a wireless-communication network. In general, the 3D antenna module may mitigate propagation losses and allow the UE to maintain a link-budget for the wireless link.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, that creates a machine learning model with k-means clustering. In some implementations, an instruction to create a model is obtained. A data set including geographic data and non-geographic data is received. The data set includes multiple data entries. Geographic centroids are determined from the geographic data. The data set is analyzed to obtain statistics of the data set. Transformed data is generated from the data set, the statistics, and the geographic centroids. A model is generated with the transformed data, the model indicating multiple data groupings.

Abstract: The present disclosure provides for using multiple inertial measurement units (IMUs) to recognize particular user activity, such as particular types of exercises and repetitions of such exercises. The IMUs may be located in consumer products, such as smartwatches and earbuds. Each IMU may include an accelerometer and a gyroscope, each with three axes of measurement, for a total of 12 raw measurement streams. A training image includes a plurality of subplots or tiles, each depicting a separate data stream. The training image is then used to train a machine learning model to recognize IMU data as corresponding to a particular type of exercise.

Abstract: Systems, methods, and media for managing an entertainment system are provided. In some implementations, systems for managing an entertainment system are provided, the systems comprising: at least one hardware processor configured to: detect a first instruction; select a component of the entertainment system; determine a first state of the component; store an indication of the first state; detect a second instruction; retrieve the indication of the first state; generate a third instruction based on the indication of the first state; and transmit the third instruction to the component of the entertainment system.

Abstract: This document describes techniques and apparatuses for a high-brightness mode on an organic light emitting diode (OLED) display. The techniques and apparatuses set a high-brightness value in a register of a display driver integrated circuit (DDIC) associated with the OLED display of on electronic device. A processor of the electronic device provides a fewer-pulses command to the DDIC, which adjusts a pulse number to control the OLED display at fewer pulses per period. A gamma correction is determined based on the high-brightness value and used to alter content to be presented on the OLED display. As a result, fewer pulses are used in combination with the gamma correction to provide content on the OLED display at a high brightness.

Abstract: This document describes techniques and systems that enable radar-based authentication status feedback. A radar field is used to enable an electronic device to account for the user's distal physical cues to determine and maintain an awareness of the user's location and movements around the device. This awareness allows the device to anticipate some of the user's intended interactions and provide functionality in a timely and seamless manner, such as preparing an authentication system to authenticate the user before the user touches or speaks to the device. These features also allow the device to provide visual feedback that can help the user understand that the device is aware of the user's location and movements. In some cases, the feedback is provided using visual elements presented on a display.

Abstract: A method for identifying insertion points for inserting live content into a continuous content stream is disclosed. The method includes identifying a first particular live content item included in a continuous content stream. The method determines a first live content item for presentation based on one or more interesting portions of the first live content item, wherein the one or more interesting portions are identified based one or more selection criteria, and wherein a first interesting portion comprises a start time. The method further causes the first live content item to be presented by the client device at the start time of the first interesting portion during playback of the first particular item.

Abstract: Systems and methods for adjusting light emitted from a display of a device are provided. The adjusting includes obtaining, from light of an environment detected by at least one sensor, a measured color of light of the environment, and obtaining, from light of the environment detected by at least one sensor, a measured brightness of light of the environment. In response to the obtaining the measured color and the measured brightness of light, a color of light emitted from the display is adjusted from an initial color prior to the adjusting to a target color that matches the measured color. Further, a brightness of light emitted from the display is adjusted from an initial brightness emitted by the display prior to the adjusting to a target brightness that matches the measured brightness of light.

Abstract: A method is implemented at an electronic device for displaying output from an application. The electronic device includes a display module and an application. The application sends to the display module a request to display output on the fixed orientation display. The display module determines whether the application is able to scale the output from the application to fit the fixed orientation display. In accordance with a determination that the application is able to scale the output, the electronic device causes the application to receive information concerning the fixed orientation display from the display module and scale the output for display on the fixed orientation display according to the information. In accordance with a determination that the application is not able to scale the output, the display module scales the output received from the application, thereby enabling the output of the application to be displayed on the fixed orientation display.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for a display with discrete gate-in-panel circuitry. In some implementations, a display includes an array of emissive pixels arranged in rows and columns, where the array includes a first continuous area having a first pixel density and a second continuous area having a second pixel density less than the first pixel density, and consecutive rows of the emissive pixels extending between the first and second continuous areas. The display also includes gate in panel (GIP) circuits in the second continuous area, a data lines connected to the array of emissive pixels, and signal lines connected to the array of emissive pixels.

Abstract: Provided is an in-ear device and associated computational support system that leverages machine learning to interpret sensor data descriptive of one or more in-ear phenomena during subvocalization by the user. An electronic device can receive sensor data generated by at least one sensor at least partially positioned within an ear of a user, wherein the sensor data was generated by the at least one sensor concurrently with the user subvocalizing a subvocalized utterance. The electronic device can then process the sensor data with a machine-learned subvocalization interpretation model to generate an interpretation of the subvocalized utterance as an output of the machine-learned subvocalization interpretation model.