Abstract: An authentication application receives an audio input, detects whether the audio input matches a known passphrase, and processes the audio input to determine whether the audio input is consistent with a known voice signature. Upon determining that the audio input is consistent with the known voice signature, the application will identify a user who is associated with the voice signature. The device will output an audio prompt, receive a spoken answer, and analyze the spoken answer to determine whether it corresponds to an expected response. Outputting the prompt may responsive to determining that the user was not authenticated to the device or the application within a threshold time period, or if a security threat was posted to the user's profile, When the system determines that the spoken answer corresponds to the expected response, it will authenticate the user and provide the user with access to the device or application.

Abstract: Users can navigate and manipulate content, such as by scrolling a document or selecting content in the document, by using various touch inputs that indicate a behavior that the user is trying to perform. In order to improve the user interface experience for a user, embodiments analyze touch inputs to determine when a user would like to perform a navigational jump and then execute such a jump, based on the specific touch inputs involved.

Abstract: Methods and systems for automatically transitioning content based on facial recognition, wherein the method may include receiving a signal from a camera, identifying a face of a viewer within an area, retrieving a user identifier associated with the face of the viewer, selecting a program, and displaying the program.

Abstract: In some implementations, authentication tokens corresponding to known users of a device are stored on the device. An utterance from a speaker is received. The utterance is classified as spoken by a particular known user of the known users. A query that includes a representation of the utterance and an indication of the particular known user as the speaker is provided using the authentication token of the particular known user.

Abstract: Methods, systems, and devices for enhanced digital headsets are disclosed. An enhanced USB-C headset includes a USB-C connector, a cable extending from the USB-C connector, an inline control box coupled to the USB-C connector through the cable, and a first earphone and a second earphone. The cable includes conductors for transmitting DC bus power, power return, and differential digital signals and extends at least one foot in length. The control box includes a single circuit board, with circuitry for managing digital communications, converting audio data, and providing output signals to drive analog speaker elements of the earphones.

Abstract: Multiple image verification challenges can be used to identify the location of an object within an initial image. For instance, a first set of tiles is generated using the initial image. This first set is provided to a client computing device for display in a first verification challenge requesting that the user select tiles including the object. In response, a user selection of tiles of the first set is received. These selected tiles are then used to generate a second set of tiles corresponding to a sub-portion of the initial image. The second set of tiles is provided to a client computing device for display in a second verification challenge. In response to the second verification challenge, a user selection of tiles of the second set is received. This user selection of tiles of the second set is then used to determine a location of the object in the image.

Abstract: Methods, systems, and media for inserting and presenting video objects linked to a source video are provided. In some embodiments, the method comprises: identifying a plurality of video objects associated with a content creator, wherein each of the plurality of video objects comprises a portion of a video associated with the content creator; causing a plurality of selectable inputs each corresponding to at least one video object to be presented by a first user device, wherein at least one of the selectable inputs corresponds to at least one of the plurality of video objects associated with the content creator; receiving an indication that one of the plurality of selectable inputs from the first user device was selected; identifying a video object to be presented by a second user device based on the received indication; and causing the identified video object to be presented by the second user device.

Abstract: A computer-implemented method for generating creative type suggestions for an online content provider is provided. The method uses a computing device including a processor and a memory. The method includes training a first model with historical information including one or more of (i) serving performance of online advertisements and (ii) advertiser information. The method also includes computing a preliminary creative type suggestion using at least the first model. The method further includes modifying the preliminary creative type suggestion based at least in part on past suggestion performance to generate a final creative type suggestion. The method also includes presenting the final creative type suggestion to the online content provider.

Abstract: A computing device may recreate data objects formatted specifically for use in a currently executing application based on other data objects formatted in a different manner and specifically for use in a secondary application. For example, the computing device may initially execute a first application. The computing device may execute a second application contains one or more data objects that are specifically formatted for use and display in the second application. The first application may receive a first data object, where a format of the first data object is specific to the second application. The first application may determine content of the first data object and create a second data object that includes at least a portion of the content of the first data object, where a format of the second data object is specific to the first application.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for switching between transmission technologies within a spectrum based on network load. In one aspect, a method includes obtaining network load information that indicates network load for a first access point operating using frequency domain multiplexing in a first spectrum and network load for a second operator using frequency domain multiplexing in a second spectrum that does not overlap any portion of the first spectrum, determining that the network load information satisfies a network load threshold, and in response to determining that the network load information satisfies the network load threshold, providing an instruction to the first access point to operate using LBT in a third spectrum the includes at least a portion of the first spectrum and at least a portion of the second spectrum.

Abstract: Network protocols generally implement integrity protection, encryption and authentication as separate validation steps. Since each validation step contributes encoding and processing overhead associated with individual packet transfers over the network, such network protocols can make inefficient use of limited packet space. Systems and methods according to the present disclosure combine integrity protection, encryption and authentication into a single validation step thereby making efficient use of limited packet space.

Abstract: Methods, systems, and apparatus are described for processing mutations for a remote database. In one aspect, a method includes receiving a log of database mutations from a client device, the log of database mutations indicating changes previously made to a client version of a database stored on the client device, and each database mutation specifying: an entity included in the database; an operation for the entity; and a client version number for the entity; identifying, in the log of database mutations, two or more database mutations that each specify a same client version number for a same entity; determining, based on a conflict resolution policy, that the two or more database mutations are eligible for collapsing; and collapsing the two or more database mutations into a single database mutation.

Abstract: The present disclosure relates to evaluating different semantic interpretations of a search query. One example method includes obtaining a set of search results for a particular search query submitted to a search engine; obtaining a set of semantic interpretations for the particular search query; obtaining, for each semantic interpretation of the set, a canonical search query; generating a modified search query based at least in part on the particular search query and the canonical search query for the semantic interpretation; obtaining a set of search results for the modified search query for the semantic interpretation; and determining, for each semantic interpretation of the set, a degree of similarity between (i) the set of search results of the modified search query for the semantic interpretation, and (ii) the set of search results for the particular search query.

Abstract: The present disclosure is directed to methods and systems for reconstructing geographic information system (GIS) scenes. In particular, the methods and systems of the present disclosure can receive data indicating a user intention to save an image from a GIS application. Responsive to receiving the data indicating the user intention to save the image, image data representing the image from the GIS application can be generated. The image data can comprise metadata configured to cause the GIS application to reconstruct a scene comprising the image.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing long-short term memory cells with saturating gating functions. One of the systems includes a first Long Short-Term Memory (LSTM) cell, wherein the first LSTM cell is configured to, for each of the plurality of time steps, generate a new cell state and a new cell output by applying a plurality of gates to a current cell input, a current cell state, and a current cell output, each of the plurality of gates being configured to, for each of the plurality of time steps: receive a gate input vector, generate a respective intermediate gate output vector from the gate input, and apply a respective gating function to each component of the respective intermediate gate output vector, wherein the respective gating function for at least one of the plurality of gates is a saturating gating function.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for adversarial training of a neural network. One of the methods includes obtaining a plurality of training inputs; and training the neural network on each of the training inputs, comprising, for each of the training inputs: processing the training input using the neural network to determine a neural network output for the training input; applying a perturbation to the training input to generate an adversarial perturbation of the training input; processing the adversarial perturbation of the training input using the neural network to determine a neural network output for the adversarial perturbation; and adjusting the current values of the parameters of the neural network by performing an iteration of a neural network training procedure to optimize an adversarial objective function.

Abstract: In some examples, a method includes, receiving, by a supervisor component, a request to execute particular functionality of an application and, responsive to receiving the request, identifying, by an application container for maintaining installable sub-packages of the application, a minimum version of a library specified in an application definition of the application. The method may also include, responsive to determining that the application container does not include the minimum version of the library: retrieving an instance of the library with a version that is greater than or equal to the minimum version; installing the instance of the library in the container runtime of the application container; and executing the application using the instance of the library in the container.

Abstract: In embodiments of distributed coordination of mesh network configuration updates, pending commissioning datasets are managed and distributed to coordinate configuration changes of parameters that control participation in, and secure communication over, a mesh network. Pending network commissioning datasets are managed across fragmentation of the mesh network into multiple partitions and subsequent merging of the fragments to ensure that the most recent updates to pending commissioning datasets are propagated to mesh network devices and that all mesh network devices will receive pending commissioning datasets before the time that the pending commissioning dataset becomes the active commissioning dataset for the mesh network.