Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for score normalization. One of the methods includes receiving initial training data, the initial training data comprising initial training records, each initial training record identifying input data as input and a category as output. The method includes generating a first trained predictive model using the initial training data and a training function. The method includes generating intermediate training records by inputting input data of the initial training records to a second trained predictive model, the second trained predictive model generated using the training function, each intermediate training record having a score. The method also includes generating a score normalization model using a score normalization training function and the intermediate training records.

Abstract: Enabling or disabling vehicle applications for execution is described. When a request to execute an application is received by a vehicle processing system or a change in the operational status of a vehicle is detected, the vehicle processing system may obtain classification data classifying the application and the current operational status of the vehicle. The classification data may indicate an application type of the application and the types of vehicle operation statuses that the application can be executed under. Based on the classification data, vehicle operation status, and one or more rules, the vehicle processing system may determine whether to enable execution of the application or deny execution of the application in the vehicle.

Abstract: An image processor is described. The image processor includes an I/O unit to read input image data from external memory for processing by the image processor and to write output image data from the image processor into the external memory. The I/O unit includes multiple logical channel units. Each logical channel unit is to form a logical channel between the external memory and a respective producing or consuming component within the image processor. Each logical channel unit is designed to utilize reformatting circuitry and addressing circuitry. The addressing circuitry is to control addressing schemes applied to the external memory and reformatting of image data between external memory and the respective producing or consuming component. The reformatting circuitry is to perform the reformatting.

Abstract: An image processor is described. The image processor includes a storage circuit to store segments of input image data received in a raster scan format from a camera. The image processor further includes a reformatting circuit to convert the segments of input image data into a block image format. The image processor further includes a processor comprising a two-dimensional execution lane array and a two-dimensional shift register array. The two-dimensional shift register array is to store the input image data that has been formatted into the block image format. The execution lane array is to execute instructions that operate on the image data from the two-dimensional shift register array.

Abstract: Certain implementations of the disclosed technology may include systems and method for handling application notifications. According to an example implementation, a method is provided. The method can include receiving a plurality of audio feeds from a respective plurality of applications, and determining a priority status for one or more of the respective applications. Based on the determined priority status, the method may further include determining a first prioritized audio feed and a second prioritized audio feed from the plurality of applications. The method includes detecting, in a signal associated with the first prioritized audio feed, a signal gap having a gap start. The method includes modifying, based at least in part on the signal of the first prioritized audio feed, one or more parameters associated with at least the second prioritized audio feed, and outputting, to an output device, at least a portion of the first prioritized audio feed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving a query provided by a user and comprising one or more terms. Obtaining context data based on at least a portion of a first resource displayed to the user at a time that the query is received. Obtaining a revised query that is based on the query and the context data. Receiving a plurality of search results responsive to the revised query. Automatically, selecting a search result that represents a second resource from the plurality of search results, and providing the second resource for display to the user.

Abstract: Instead of accepting competing ads and using an arbitration function (e.g., an auction) to choose winning ads to be served with a document, sets of ads (perhaps having different characteristics) can be generated, and an arbitration function can be used to select the winning set of ads. Such arbitrations on sets of ads can consider how ads, search results, colors, positions, fonts, etc., all interact with each other and affect the usefulness of the sets of ads to advertisers, end users, document publishers, and/or an ad serving entity.

Abstract: Systems and methods for distributing network traffic during streaming are hereby provided. A server can provide a first content pod to a client device for insertion in a content break separating a first portion and a second portion of the streaming content. The server can receive, from the client device during playback of the first content pod, a request for content to insert into the content break for playback subsequent to playback of the first content pod. The request for content can be generated at a first elapsed time from initiation of playback of the first content pod via a request generation policy. The server can select a second content pod. The server can provide the second content pod to the client device during the playback of the first content pod for playback subsequent to completion of the playback of the first content pod.

Abstract: This document describes mobility management of edge computing resources in fifth generation new radio (5G NR) wireless networks. The techniques described enable authorizing user devices to access edge compute servers that execute applications for the user device. The techniques described also enable the migration of applications of user devices between edge compute servers based on mobility changes of user devices in a wireless network, such as handovers of a user device between base stations in the wireless network.

Abstract: A language translation application on a user device includes a user interface that provides relevant textual and graphical feedback mechanisms associated with various states of voice input and translated speech.

Abstract: Methods, systems, and media for controlling access to content are provided. In some embodiments, the method comprises: receiving a request to modify a first access list associated with a container item, wherein the first access list indicates a set of users that are authorized to access the container item; identifying the first access list in response to the request; accessing a second access list associated with a content item embedded in the container item, wherein the second access list indicates a set of users that are authorized to access the content item; modifying the first access list to filter out users from the first access list that are not included in the second access list; and storing the first access list such that the first access list is accessible to a host of the container item to verify that a requesting user is authorized to access the container item.

Abstract: In general, the subject matter described in this specification can be embodied in methods, systems, and program products for receiving user input that defines a search query, and providing the search query to a server system. Information that a search engine system determined was responsive to the search query is received at a computing device. The computing device is identified as in a first state, and a first output mode for audibly outputting at least a portion of the information is selected. The first output mode is selected from a collection of the first output mode and a second output mode. The second output mode is selected in response to the computing device being in a second state and is for visually outputting at least the portion of the information and not audibly outputting the at least portion of the information. At least the portion of information is audibly output.

Abstract: A wireless device and method detect a current signal strength of each of a number of wireless networks, determine an expected end-to-end data speed for each of the wireless networks using wireless network end-to-end data speed information that was historically compiled from prior connections with the wireless networks and using the detected current signal strength of each of the wireless networks. The wireless device and method connects to at least one wireless network based on the determined expected end-to-end data speed and/or provides a user interface that includes the determined expected end-to-end data speed and the detected signal strength indication of each of the wireless networks.

Abstract: Systems and methods of a gain adjustment component for content selection infrastructure are provided. The system can receive a selection of multiple topics identified by a semantic analysis technique, and identify one or more keywords. The system can determine relevance scores between each of the keywords and the multiple topics. The system can generate an aggregated relevance score for each keyword based on the relevance score for the keyword and each of the multiple topics. The system can determine a threshold based on a metric associated with the multiple topics. The system can determine to input or exclude each of the one or more keywords based on a comparison of the corresponding aggregated relevance score with the threshold.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing customized television advertisements. In one aspect, television advertising templates are used for generating customized television advertisements. The television advertisement templates include targeting criteria specifying targeting conditions for the television advertising template, which are conditions for selecting the television advertisement template for an advertisement availability, and content selection criteria specifying availability content associated with an advertisement availability for use in selecting variable advertisement content elements for inclusion in a customized television advertisement generated from the television advertising template. The variable advertisement content elements can include video elements, audio elements, and text elements.

Abstract: Methods, systems, and apparatus, including an apparatus for transferring data using multiple buffers, including multiple memories and one or more processing units configured to determine buffer memory addresses for a sequence of data elements stored in a first data storage location that are being transferred to a second data storage location. For each group of one or more of the data elements in the sequence, a value of a buffer assignment element that can be switched between multiple values each corresponding to a different one of the memories is identified. A buffer memory address for the group of one or more data elements is determined based on the value of the buffer assignment element. The value of the buffer assignment element is switched prior to determining the buffer memory address for a subsequent group of one or more data elements of the sequence of data elements.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for dismissing displayed elements. In some implementations, a device with a touchscreen is operated using a first touch input threshold value, which may correspond to a minimum distance of input to the touchscreen to identify a gesture. An automobile mode event is detected that initiates an automobile mode for the device. A user input to the touchscreen of the device is received while the device is in the automobile mode. Using a second touch input threshold value corresponding to the automobile mode, the device determines that the user input to the touchscreen of the device represents a gesture corresponding to a user interface element displayed on the device. The device dismisses the user interface element so that the user interface element is no longer displayed.

Abstract: A computer-implemented method for speech diarization is described. The method comprises determining temporal positions of separate faces in a video using face detection and clustering. Voice features are detected in the speech sections of the video. The method further includes generating a correlation between the determined separate faces and separate voices based at least on the temporal positions of the separate faces and the separate voices in the video. This correlation is stored in a content store with the video.

Abstract: Methods, systems, and media for identifying content in stereoscopic videos and, more particularly, for detecting abusive stereoscopic videos by generating fingerprints for multiple portions of a video frame are provided.

Abstract: Video segments related to an annotation term are identified from a target video. A video dataset and an image data set are searched using the annotation term to generate a video set and an image set. The video set and the image set are iteratively refined to generate a set of iconic images. A frame level model is generated using the set of iconic images and video segments related to the annotation term are identified from the target video by applying the frame level model to frames of the target video.