Abstract: Techniques are described herein for detecting and/or enrolling (or commissioning) new “hot commands” that are useable to cause an automated assistant to perform responsive action(s) without having to be first explicitly invoked. In various implementations, an automated assistant may be transitioned from a limited listening state into a full speech recognition state in response to a trigger event. While in the full speech recognition state, the automated assistant may receive and perform speech recognition processing on a spoken command from a user to generate a textual command. The textual command may be determined to satisfy a frequency threshold in a corpus of textual commands. Consequently, data indicative of the textual command may be enrolled as a hot command. Subsequent utterance of another textual command that is semantically consistent with the textual command may trigger performance of a responsive action by the automated assistant, without requiring explicit invocation.

Abstract: The subject matter described herein provides systems and techniques for a network planning and optimization tool that may allow for network capacity planning using key network failures for an arbitrary pair of network topology and demands. Performing network capacity planning with key network failures, instead of using other techniques, may avoid over-building the topology of a network. In particular, key network failures may be generated from the probabilistic failures, and the impact of these failures on a network may be computed. Expected flow availability SLO or a function thereof may be computed, using this information, and used by the tool to design a robust network. With an embedded flow availability calculation and updated risk framework, the capacitated cross-layer network topologies output by the tool may meet network demands/flows with their respective SLO type at the lowest cost.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating an output sequence from an input sequence. In one aspect, one of the systems includes an encoder neural network configured to receive the input sequence and generate encoded representations of the network inputs, the encoder neural network comprising a sequence of one or more encoder subnetworks, each encoder subnetwork configured to receive a respective encoder subnetwork input for each of the input positions and to generate a respective subnetwork output for each of the input positions, and each encoder subnetwork comprising: an encoder self-attention sub-layer that is configured to receive the subnetwork input for each of the input positions and, for each particular input position in the input order: apply an attention mechanism over the encoder subnetwork inputs using one or more queries derived from the encoder subnetwork input at the particular input position.

Abstract: This technology relates to a destruction validation system. The destruction validation system includes a destruction device and a controller. The controller is configured to monitor current drawn by the destruction device during destruction of the piece of physical media. The controller may compare the current drawn to a current profile. Based on the comparison of the current drawn to the current profile, the controller may generate and output a notification indicating whether the piece of physical media has been destroyed. In some instances, the destruction device includes a current sensing device. The current sensing device may be configured to detect the current drawn by the destruction device and transmit the detected current drawn to the controller. The current drawn that is monitored by the controller may be the detected current drawn by the current sensing device.

Abstract: A method includes receiving a reference audio signal corresponding to reference speech spoken by a target speaker with atypical speech, and generating, by a speaker embedding network configured to receive the reference audio signal as input, a speaker embedding for the target speaker. The speaker embedding conveys speaker characteristics of the target speaker. The method also includes receiving a speech conversion request that includes input audio data corresponding to an utterance spoken by the target speaker associated with the atypical speech. The method also includes biasing, using the speaker embedding generated for the target speaker by the speaker embedding network, a speech conversion model to convert the input audio data corresponding to the utterance spoken by the target speaker associated with atypical speech into an output canonical representation of the utterance spoken by the target speaker.

Abstract: A method for a virtual network of a mobile device includes, for the mobile device executing a user application and simultaneously connected to a first wireless network and a second wireless network, selecting a remote server from a plurality of remote servers. The method also includes establishing the virtual network between the mobile device and the selected remote server. The virtual network includes at least one tunnel between the mobile device and the selected remote server over the first wireless network and the second wireless network. The method also includes transmitting data from the user application to the selected remote server over the virtual network. The data, when received by the selected remote server, causes the selected remote server to route the data to a destination server.

Abstract: A method includes determining a cluster reliability of a computing cluster including a maximum computing capacity and representative of a reliability of the computing cluster when utilizing an entirety of the maximum computing capacity. The operations include receiving a provisioning request of the computing cluster including a threshold reliability of the computing cluster. In response to the provisioning request, determining, using the cluster reliability, a reserved computing capacity of the computing cluster based on the threshold reliability. The reserved computing capacity is less than the maximum computing capacity. Based on the reserved computing capacity and the maximum computing capacity, the operations include determining an unreserved computing capacity of the computing cluster. The operations include provisioning the computing cluster for execution of a user workload. The user workload executes on the unreserved computing capacity.

Abstract: This document describes techniques and devices for blood-solute calculation with a mobile device using non-invasive spectroscopy. A mobile device includes a light source that emits light toward an interferometer that uses mirrors to separate and recombine the light. The interferometer directs the recombined light toward a person. Light reflected from, or transmitted through, the person is received through a reception port to a photodetector that outputs photodetector data that corresponds to a measured light intensity of the reflected and transmitted light as a function of a path length of the light or a mirror position of the interferometer. Based on the photodetector data, an interferogram is generated. Applying a technique such as a Fourier transform to the interferogram, a spectrum data set of the reflected and transmitted light is generated. Based on the spectrum data set, a concentration of solutes in the person's blood is calculated.

Abstract: A method includes assigning a plurality of workloads to an isolated secure compute environment on a distributed computing system based on respective security criteria for each of the plurality of workloads, the isolated secure compute environment isolated from one or more other isolated secure compute environments on the distributed computing system. While executing the plurality of workloads in the isolated secure compute environment on the distributed computing system, the method also includes determining resource utilization for the isolated secure compute environment, and adjusting, using the determined resource utilization, a number of computing resources assigned to the isolated secure compute environment.

Abstract: A method for providing elastic columnar cache includes receiving cache configuration information indicating a maximum size and an incremental size for a cache associated with a user. The cache is configured to store a portion of a table in a row-major format. The method includes caching, in a column-major format, a subset of the plurality of columns of the table in the cache and receiving a plurality of data requests requesting access to the table and associated with a corresponding access pattern requiring access to one or more of the columns. While executing one or more workloads, the method includes, for each column of the table, determining an access frequency indicating a number of times the corresponding column is accessed over a predetermined time period and dynamically adjusting the subset of columns based on the access patterns, the maximum size, and the incremental size.

Abstract: A method for on demand serverless container based storage transfer includes receiving a request to transfer data from a first device to a second device, the first device hosted at a private cloud, the private cloud isolated from the Internet. The method includes determining that the first device is communicatively connected to the private cloud. The method also includes, in response to determining that the first device is communicatively connected to the private cloud, instantiating a container at the first device, the container configured to receive the data from the first device without directly accessing a local storage of the first device. The method includes transferring, using the container, the data from the first device to the second device.

Abstract: Various arrangements of antenna systems are detailed herein. Such an antenna system can include first antenna connected to a printed circuit board. The antenna system further includes a second antenna connected to the printed circuit board. The antenna system further includes a decoupling structure connected to the printed circuit board at a location between the first antenna and the second antenna.

Abstract: This document discloses techniques, apparatuses, and systems for efficient and secure data handling using domain-oriented masking. In aspects, efficient and secure data handling using domain-oriented masking enables circuit resources (e.g., flip flops, data) to be shared between portions of an integrated circuit (IC). Specifically, an input register of a Galois-Field (GF) multiplier and a pipeline register within a stage of an Advanced Encryption Standard Substitution-Box (S-Box) may be implemented as sharing a single flip flop. Some ICs may utilize multiple S-Boxes, including multiple GF multipliers that are provided randomness each time the input is updated. In this case, an output from a previous stage of a first S-box may be provided as randomness to a subsequent stage of a second S-Box to limit the need for entropy generation circuitry. In this way, efficient and secure data handling using domain-oriented masking may decrease circuit cost and circuit area without impacting security.

Abstract: This document describes systems and techniques for determining whether an earbud is removed from within an ear of a user. The systems and techniques employ an earbud including a housing having a distal end and a proximal end. The distal end includes a magnetically-sensitive section, the distal end of the housing being configured to be magnetically attracted to a charging receptacle and insertable within an ear of a user. On-head detection (OHD) logic is electrically coupled to the magnetically-sensitive section at the distal end, the OHD logic configured to determine based on a distal-end capacitance measured using the magnetically-sensitive section whether the distal end of the housing is within the ear of the user.

Abstract: A method for in process asynchronous out of memory logs for remote containerized deployments includes executing a container process within a container. The method further includes, writing, by the container process, a log of the container process to a first log file. The method also includes storing, by the container process, the first log file at non-volatile memory mounted to the container. The method includes determining, by the container process, that the first log file satisfies a threshold size. In response to determining that the first log file satisfies the threshold size, the method includes writing the log of the container process to a second log file, compressing the first log file into a first compressed log file, and transmitting the first compressed log file to a remote endpoint.

Abstract: The technology provides a model-based approach for multilingual text rewriting that is applicable across many languages and across different styles including formality levels or other textual attributes. The model is configured to manipulate both language and textual attributes jointly. This approach supports zero-shot formality-sensitive translation, with no labeled data in the target language. An encoder-decoder architectural approach with attribute extraction is used to train rewriter models that can thus be used in “universal” textual rewriting across many different languages. A cross-lingual learning signal can be incorporated into the training approach. Certain training processes do not employ any exemplars. This approach enables not just straight translation, but also the ability to create new sentences with different attributes.

Abstract: A headset in the form of glasses for use in creating an augmented reality experience is implemented with a capacitive smart sensor disposed on an inside surface of an arm of the glasses. The capacitive smart sensor can be trained, using a neural network, to identify and distinguish different gestures associated with a touch event on an outside surface of the arm of the glasses. The smart sensor thus transforms the headset into a user input device without adding a touch pad.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for query categorization based on image results. In one aspect, a method includes receiving images from image results responsive to a query, wherein each of the images is associated with an order in the image results and respective user behavior data for the image as a search result for the first query, and associating one or more of the first images with a plurality of annotations based on analysis of the selected first images' content.

Abstract: A system for controlling an impedance matching circuitry can include an antenna, a receiving or transmitting circuitry, an impedance matching circuitry, a circuitry, and a controller. The impedance matching circuitry can be coupled between the antenna and the receiving or transmitting circuitry. The circuitry can be configured to set a value of an impedance of the impedance matching circuitry. The controller can be configured to determine, based on a state of an electronic device, that the electronic device is configured to perform one or more functions with one or more of a satellite radio navigation signal at a first or a second frequency, a wireless local area network signal at a third or a fourth frequency, or a personal area network signal at a fifth or a sixth frequency. The controller can be configured to control, in response to a determination of the state of the electronic device, the circuitry.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing conversion path performance measures and reports. In one aspect, user interaction data are obtained, were the user interaction data specify user interactions for a plurality of conversions. User interactions that are associated with each conversion are selected from the user interaction data, where the associated user interactions for each conversion are user interactions with a converting user during the conversion cycle for the conversion. Using the user interaction data for the selected user interactions, a quantity of user interactions that are associated with each conversion and occurred during the conversion cycle for the conversion are determined. In turn, conversion path performance measures are computed and reports specifying the conversion path performance measures are generated.