Abstract: A method for instantaneous learning in text-to-speech (TTS) during dialog includes receiving a user pronunciation of a particular word present in a query spoken by a user. The method also includes receiving a TTS pronunciation of the same particular word that is present in a TTS input where the TTS pronunciation of the particular word is different than the user pronunciation of the particular word. The method also includes obtaining user pronunciation-related features and TTS pronunciation related features associated with the particular word. The method also includes generating a pronunciation decision selecting one of the user pronunciation or the TTS pronunciation of the particular word that is associated with a highest confidence. The method also include providing the TTS audio that includes a synthesized speech representation of the response to the query using the user pronunciation or the TTS pronunciation for the particular word.

Abstract: Systems and methods are provided for acquiring training data and building an organizational-based language model based on the training data. In organizational data is generated via one or more applications associated with an organization, the collected organizational data is aggregated and filtered into training data that is used for training an organizational-based language model for speech processing based on the training data.

Abstract: Automated monitoring of a voice communication session, when the session is in an on hold status, to determine when the session is no longer in the on hold status. When it is determined that the session is no longer in the on hold status, user interface output is rendered that is perceptible to a calling user that initiated the session, and that indicates that the on hold status of the session has ceased. In some implementations, an audio stream of the session can be monitored to determine, based on processing of the audio stream, a candidate end of the on hold status. In response, a response solicitation signal is injected into an outgoing portion of the audio. The audio stream can be further monitored for a response (if any) to the response solicitation signal. The response (if any) can be processed to determine whether the end of the on hold status is an actual end of the on hold status.

Abstract: One implementation of a sensing method includes: correlating a receive signal with a waveform template to obtain a correlation signal in which echoes would be represented as peaks; and varying the waveform template based on time elapsed from a measurement start time.

Abstract: A securing apparatus configured to assist in securing objects to a pallet during packaging and shipping includes a main body having first and second wall portions spaced from one another to define a channel, and a hook extending from the main body via a neck portion. At least one of the first and second wall portions includes an anchor for engaging and being embedded in a pallet. A fastener may be received through an aperture of the main body to releasably attach the securing apparatus to a pallet, and the anchor may be at least partially embedded in the pallet. So configured, a filament may be looped about the hook to assist in securing objects to the pallet in a first securing configuration, and another filament may be received in the channel to assist in securing objects to the pallet in a second securing configuration.

Abstract: Thermal management systems for batteries utilize vapor chambers having wicking components therein. An exemplary thermal management system includes a vapor chamber containing a working fluid and wicking components. A plurality of battery cells are disposed at least partially in the vapor chamber. A cold plate is coupled to the vapor chamber, and a heat pump is coupled to the cold plate. A capillary tube may be utilized to facilitate movement of vapor and working fluid in the thermal management system. Via use of exemplary systems, improved thermal management for batteries is provided.

Abstract: A hydraulic fracturing system has a pump driven by an electrically powered motor. The pump pressurizes fluid which is piped into a wellbore to fracture a subterranean formation. The pump and motor are mounted on a trailer that is hitched to a tractor. A platform assembly is mounted onto the trailer, and which is selectively moveable between deployed and stowed configurations. The platform assembly includes a platform, a lateral rail assembly mounted to the platform, and gates on the forward and aft ends of the platform. The rail assembly and gates define a safety barrier to prevent operations personnel from falling off the platform. In the stowed configuration the platform assembly is anchored in place over wheels on the trailer. In the deployed configuration, the platform assembly provides work surface so that operations personnel can readily access the pump on the trailer.

Abstract: Example methods disclosed herein to initialize classification vectors of a neural network include ranking a plurality of classes to be represented by the classification vectors based on respective numbers of instances of training data associated with corresponding ones of the classes. Disclosed example methods also include initializing the classification vectors to span a classification space corresponding to the classes. Disclosed example methods further include assigning respective ones of the classes to corresponding ones of the classification vectors based on the ranking of the classes.

Abstract: A method includes, based on comparing a velocity, lane selection and path selection of a target vehicle with respect to a host vehicle, determining a following risk assessment for the target vehicle relative to the host vehicle, determining the target vehicle is following the host vehicle based on the following risk assessment being above a threshold, and then transmitting data identifying the target vehicle to a remote computer and actuating vehicle components to operate the host vehicle to a location specified by the remote computer.

Abstract: Systems, methods, and computer-readable storage media for generating a quantifiable communication profile of a participant, then using that communication profile to identify fake media and/or verify media as authentic. An exemplary system can receive communications in which a participant was one of the parties communicating, perform natural language processing and statistical analyses on the received communications, and generate a participant communication profile which details how the participant communicates and prefers to be communicated with. The system can also receive media purported to be from the participant, perform natural language processing on the purported media, compare the syntax of the purported media to the participant's communication profile, and identify, based on word usage, the media as a fake.

Abstract: A latex emulsion may include an aqueous carrier liquid and a latex copolymer formed from reactants comprising butyl methacrylate, wherein the reactants include at least 20 wt. % butyl methacrylate, based on the total weight of ethylenically unsaturated monomers used to make the latex copolymer. A coating formed from the latex emulsion exhibits a wet adhesion to a thermoplastic polyolefin roofing membrane of greater than about 1 pound per linear inch when tested using fabric embedded peel adhesion testing per ASTM C794. The latex emulsion may be used as part of an aqueous coating composition or a roofing system including a polymeric roofing membrane.

Abstract: Systems, methods, computer program products, and devices provide for computing an eigensystem from a first data set; computing updated eigenvalues that approximate an eigensystem of at least a second data set based on the eigensystem of the first data set; and evaluating a plurality of features in each of the first and at least second data sets using a cost function. The cost function uses fewer than the total number of eigenvalues and can include a condition number. The cost function can perform a coarse approximation of the eigenvalues to de-select at least one of the data sets. This can be useful for learning and/or online processing in an artificial neural network.

Abstract: Methods, systems, and computer readable media can be operable to facilitate an exchange of messages between an access point and a station, wherein the access point requests a unique identifier from the station. The station may either respond with a message declining to provide a unique identifier or respond with a message including a unique identifier to be used by the access point for the station. The response from the station may include additional limitations on the use of the unique identifier by the access point. The access point may enforce different policies against a station depending upon how the station responds to the unique identifier request.

Abstract: A hotel or motel room incorporates special effects in accordance with a particular theme (e.g., haunted room). A control room provides a concealed area for personnel and/or automated systems to trigger the special effects which may include, but are limited to shaking, rattling, rumbling, pepper ghosts, water fixtures being activated, thunder and lightning, swapping tiles with digital graphics, turning on a TV, clock or light, etc. Heat sensors, weight sensors, floor mats and/or other sensors determine positions of guests in the hotel and hotel rooms thereby allowing the automatic or manual triggering of the special effects.

Abstract: A semiconductor structure includes a memory die bonded to a logic die. The memory die includes an alternating stack of insulating layers and electrically conductive layers; memory openings extending through the alternating stack, memory opening fill structures located in the memory openings and comprising a respective vertical semiconductor channel and a respective memory film, a source layer contacting the vertical semiconductor channels, a backside isolation dielectric layer contacting a backside surface of the source layer, and a source power supply mesh including a planar portion of a source-side electrically conductive layer that is located on a backside of the backside isolation dielectric layer and electrically connected to the source layer by conductive material portions that extend through the backside isolation dielectric layer.

Abstract: The present invention provides for the integration of drug dispersion methods into a drug or medicine delivery system. The drug dispersion methods used include shear (e.g., air across a drug, with or without a gas assist), capillary flow or a venturi effect, mechanical means such as spinning, vibration, or impaction, and turbulence (e.g., using mesh screens, or restrictions in the air path). These methods of drug dispersion allow for all of the drug in the system to be released, allowing control of the dosage size. These methods also provide for drug metering, fluidization, entrainment, deaggragation and deagglomeration. The present invention also provides for the integration of a drug sealing system into the device. The drug sealing system provides a way of blocking the migration of drug from one area of the package to another.

Abstract: In a general aspect, a quantum logic gate is performed in a quantum computing system. In some cases, a pair of qubits are defined in a quantum processor; the pair of qubits can include a first qubit defined by a first qubit device in the quantum processor and a second qubit defined by a tunable qubit device in the quantum processor. A quantum logic gate can be applied to the pair of qubits by communicating a control signal to a control line coupled to the tunable qubit device. The control signal can be configured to modulate a transition frequency of the tunable qubit device at a modulation frequency, and the modulation frequency can be determined based on a transition frequency of the first qubit device.

Abstract: The techniques described herein enable a head-mounted display device to use a fiducial marker to identify an Internet of Things (IoT) device. The head-mounted display device can use the identifier to establish a network connection with the IoT device. For example, the identifier can include an Internet Protocol (IP) address, a Bluetooth address, a cloud IoT identifier (e.g., AZURE hub IoT identifier), or another type of an identifier. By using an electronic paper display, the IoT device can dynamically generate and display a new fiducial marker when a new identifier is assigned to the IoT device or is generated by the IoT device. Consequently, the head-mounted display device can detect the fiducial marker and extract the identifier for the IoT device from the fiducial marker so that the identifier can be used to establish a network connection with the IoT device.