Abstract: The subject matter of this specification can be embodied in, among other things, a method that includes receiving an audio signal and initiating speech recognition tasks by a plurality of speech recognition systems (SRS's). Each SRS is configured to generate a recognition result specifying possible speech included in the audio signal and a confidence value indicating a confidence in a correctness of the speech result. The method also includes completing a portion of the speech recognition tasks including generating one or more recognition results and one or more confidence values for the one or more recognition results, determining whether the one or more confidence values meets a confidence threshold, aborting a remaining portion of the speech recognition tasks for SRS's that have not generated a recognition result, and outputting a final recognition result based on at least one of the generated one or more speech results.

Abstract: Systems and methods of automatically detecting styles of properties of DOM elements of an information resource are disclosed. A data processing system including one or more processors can identify an information resource for which to automatically detect styles of properties used to style primary content of the information resource, identify at least one DOM container that contains at least a portion of the primary content, determine, for each of one or more DOM elements included in the DOM container, styles of properties corresponding to a respective DOM element, store, in a memory, the determined styles of properties corresponding to the respective DOM elements included in the DOM container and use the stored styles of properties to modify third-party content to be inserted within the information resource.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for natural language processing. One of the method includes receiving a voice input from a user device; generating a recognition output; receiving a user selection of one or more terms in the recognition output; receiving a user input of one or more letters replacing the user selected one or more terms; determining suggested correction candidates based in part on the user input and the voice input; and providing one or more suggested correction candidates to the user device as suggested corrected recognition outputs.

Abstract: A method for providing access to dynamically provisioned resources and on-demand courses includes receiving, by an instantiation engine executing on a first computing device, from a second computing device associated with a user, an indication that the user has purchased access to the course at a scheduled time. A type of computing resource is identified for use during the course at the scheduled time. A service provider computing device is selected from a plurality of service provider computing devices, based upon an ability to provide access to the identified type of computing resource at the scheduled time. Before the scheduled time, the instantiation engine instructs the selected service provider computing device to provision a computing resource of the identified type. The method includes transmitting, to the second computing device, an instruction to establish a connection to the provisioned computing resource at the scheduled time.

Abstract: Implementations can relate to providing computer-assisted text and visual styling for images. In some implementations, a computer-implemented method includes determining a set of characteristics of an image, and applying one or more first visual modifications to the image based on one or more of the set of characteristics of the image. The method can include receiving user input defining user text, providing the user text in the image, and applying one or more second visual modifications to the image based on the user text and based on at least one of the set of characteristics of the image.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium for selecting content. A method includes: receiving an initial request for content that is to be presented along with search results responsive to a search request; identifying one or more eligible content items from the candidate content items that satisfy the initial request; receiving information describing the search results; re-processing the initial request using the information including one or more of adding other eligible content items, removing one or more previously eligible content items, adjusting a quality score or a bid associated with one or more eligible content items, or re-ranking the eligible content items; conducting an auction based on ranked eligible content items; and selecting one or more content items from among the eligible content items based on the results of the auction and transmitting the selected one or more content items.

Abstract: The disclosure is directed to maintaining an asynchronous communication via an intermediary. An agent executed on a device intermediary to clients and providers can receive a request from a client for an asynchronous communication between the client device and a provider device. The agent identifies a provider dynamic identifier identifying the provider. The agent uses a communication channel selection policy to determine a communication channel and a provider static identifier corresponding to the provider device. The agent maps a client dynamic identifier assigned by the server to a client static identifier of the client device. The agent generates a transmission data package including the content, a source address identifying the client dynamic identifier and a destination address identifying the provider static identifier.

Abstract: This specification describes technologies relating to inferring interactions with advertisers. In general, one aspect of the subject matter described in this specification can be embodied in methods that include providing an online advertisement to a system, receiving data from the system indicating an interaction of the system with the online advertisement and data indicating a subsequent interaction of the system with an online resource of an advertiser associated with the online advertisement, and inferring that an advertisement conversion has occurred based on the received data. Other embodiments of this aspect include corresponding systems, apparatus, and computer program products.

Abstract: This document describes interactive cords. An interactive cord includes a cable, and fabric cover that covers the cable. The fabric cover includes one or more conductive threads woven into the fabric cover to form one or more capacitive touchpoints which are configured to enable reception of touch input that causes a change in capacitance to the one or more conductive threads. A controller, implemented at the interactive cord or a computing device coupled to the interactive cord, can detect the change in capacitance and trigger one or more functions associated with the one or more capacitive touchpoints. For example, when implemented as a cord for a headset, the controller can control audio to the headset, such as by playing the audio, pausing the audio, adjusting the volume of the audio, skipping ahead in the audio, skipping backwards in the audio, skipping to additional audio, and so forth.

Abstract: Techniques include outputting to a developer an offer to opt-in to a translation feature that enables human translators to translate their web page to a target language. In response to receiving a first request to opt-in to the translation feature, the server: generates and stores a web page copy, obtains from the human translators translations of at least a portion of the web page from its source language to the target language, modifies the web page copy based on the obtained translations to obtain a translated web page that is a translated version of the web page, detects a second request for the web page from a computing device associated with the target language, and in response to detecting the second request outputs, to the computing device, the translated web page with additional content relevant to the computing device or a user associated with the computing device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for voice recognition. In one aspect, a method includes the actions of receiving a voice input; determining a transcription for the voice input, wherein determining the transcription for the voice input includes, for a plurality of segments of the voice input: obtaining a first candidate transcription for a first segment of the voice input; determining one or more contexts associated with the first candidate transcription; adjusting a respective weight for each of the one or more contexts; and determining a second candidate transcription for a second segment of the voice input based in part on the adjusted weights; and providing the transcription of the plurality of segments of the voice input for output.

Abstract: A method of adding a user account to an unassociated device may include detecting, by an associated device that is associated with a user account, an audio signal broadcast by an unassociated device that is not associated with the user account, where the audio signal includes a token. The method may include sending the token to a computing device associated with a service provider of the user account, receiving, by the associated device, a command, determining that the received command is an authorization command, and in response to determining that the received command is an authorization command, sending one or more authorization instructions to the computing device. The one or more authorization instructions may instruct the computing device to send one or more credentials associated with the user.

Abstract: In some implementations, data indicating a touch received on a proximity-sensitive display is received while the proximity-sensitive display is presenting one or more items. In one aspect, the techniques describe may involve a process for disambiguating touch selections of hypothesized items, such as text or graphical objects that have been generated based on input data, on a proximity-sensitive display. This process may allow a user to more easily select hypothesized items that the user may wish to correct, by determining whether a touch received through the proximity-sensitive display represents a selection of each hypothesized item based at least on a level of confidence that the hypothesized item accurately represents the input data.

Abstract: Techniques for merging multiple filters into a single GPU program is described. The system includes a filtering engine, which receives an input to apply a plurality of filters to a source image. The filtering engine identifies a first type of filter and a second type of filter from the input. The filtering engine identifies a supplemental transformation filter from the input, implements the supplemental transformation filter using a custom function to generate a color value at the source image and merges the first type of filter and the second type of filter based on the supplemental transformation filter. Finally, the filtering engine may apply the merged filter on the source image to generate a destination image.

Abstract: A method receiving an input signal at radio circuitry, sampling the input signal, and determining a power level of the sampled input signal. The radio circuitry includes an input switch having an input, a first output, and a second output. The input switch is configured to switch between the first output for the receive mode and the second output for the transmit mode. The method also includes determining whether the power level of the sampled input signal is greater than a threshold power level. When the power level of the sampled input signal is greater than the threshold power level, the method includes switching the input switch to the second output for the transmit mode. When the power level of the sampled input signal is less than or equal to the threshold power level, the method includes switching the input switch to the first output for the receive mode.

Abstract: Accessing an energy management policy for a plurality of devices is described, wherein the devices are coupled with a first structure. The energy usage of the devices is monitored. An energy usage rule and energy usage is then compared. The energy management policy and energy usage is also compared. Based on the comparing, an instruction is generated to modify an energy usage profile of said device to correlate with the energy usage rule associated with the devices and the energy management policy, thereby enabling efficient energy management.

Abstract: A method of verifying an account login request may include identifying one or more Internet Protocol (IP) addresses as potentially malicious and, for each identified IP address, determining a first value equal to a number of times over a period of time that the IP address was identified by the computing device as potentially malicious, determining a second value equal to a number of times over the period of time that the IP address was confirmed to be potentially malicious by one or more account users, determining a proportion value associated with the IP address by determining a ratio of the second value to the first value, labeling the IP address as safe or unsafe based on at least the proportion value associated with the IP address, and maintaining, by the computing device, a list of the IP addresses and corresponding labels.

Abstract: A content management computing device for selecting an online content item for display on a client computing device is provided. Additionally, a method for selecting an online content item for display on a client computing device is provided. In addition, a computer-readable storage device having processor-executable instructions embodied thereon for selecting an online content item for display on a client computing device is provided.

Abstract: Implementations generally relate to increasing engagement among users in a social network system. In some implementations, a method includes determining at least one social activity associated with an initiator user in a social network system. The method also includes enabling a token to be sent from an initiator device to one or more nearby devices, wherein the initiator device is associated with the initiator user, and wherein the one or more nearby devices are associated with one or more nearby users. The method also includes sending one or more invitations to one or more of the nearby users in association with the token. The method also includes performing one or more predetermined actions in association with the one or more nearby users based on one or more responses to the invitations.

Abstract: A method may include identifying, from a set of applications, a subset of the set of applications, each application from the subset of the set of applications being predicted, by a computing device, to be selected by a user. The method may also include outputting a graphical user interface that includes: a plurality of application icons representing the set of applications and positioned around at least a portion of a perimeter of the graphical user interface; and a plurality of prediction icons positioned within an interior of the graphical user interface and representing the subset of the set of applications. The position of a particular prediction icon representing a particular application may be based on a position of a particular application icon representing the particular application. The method may further include executing an action associated with the particular prediction icon or the one of the plurality of application icons.