Abstract: A server allows pairing of a computing device with a companion computing device so that they may automatically connect to one another in the future. Pairing is accomplished by responding to a user request to pair a particular companion device by identifying computing devices associated with a particular user account, and directing those devices to monitor whether they are in range of a proximity signal (such as a BLUETOOTH® signal). A user interface on the companion device displays a list of devices specific to the user that are in proximity to the companion computing device, so that the user can choose a device to pair with the companion computing device.

Abstract: Example speech enhancement systems include a spatio-temporal residual network configured to receive video data containing a target speaker and extract visual features from the video data, an autoencoder configured to receive input of an audio spectrogram and extract audio features from the audio spectrogram, and a squeeze-excitation fusion block configured to receive input of visual features from a layer of the spatio-temporal residual network and input of audio features from a layer of the autoencoder, and to provide an output to the decoder of the autoencoder. The decoder is configured to output a mask configured based upon the fusion of audio features and visual features by the squeeze-excitation fusion block, and the instructions are executable to apply the mask to the audio spectrogram to generate an enhanced magnitude spectrogram, and to reconstruct an enhanced waveform from the enhanced magnitude spectrogram.

Abstract: An electronic device comprises a haptic feedback system. The haptic feedback system includes a first haptic actuator coupled to a controller via a first set of two or more electrodes, and a second haptic actuator coupled to the first haptic actuator, and further coupled to the controller via a second set of two or more electrodes. The controller is configured to provide a first drive signal to the first haptic actuator via the first set of two or more electrodes, and to provide a second drive signal, different from the first drive signal, to the second haptic actuator via the second set of two or more electrodes. Combining two haptic actuators allows for a broader range of feedback and a heightened user experience.

Abstract: Techniques for prioritization of data streams include transmitting a plurality of data streams to a first device over a network where each data stream is configured to cause the first device to perform a specific function, dynamically detecting a bandwidth limitation limiting an amount of available bandwidth for transmitting the plurality of data streams, dynamically determining bandwidth reduction criteria by identifying a first subset of the plurality of data streams to stop transmitting based on the specific function that each respective data stream causes the first device to perform, temporarily stopping transmitting of the first subset of data streams configured to cause the first device to perform a first set of functions, and continuing to transmit a second subset of data streams not included in the first subset of the data streams to the first device to cause the first device to continue to perform a second set of functions.

Abstract: A server system is provided that includes a platform server system configured to store platform map data that is geospatially referenced and includes building outline data for one or more buildings, and provide the platform map data to client computer devices. The server system further includes a tenant bounded server system including one or more processors configured to store building floorplan data for a tenant entity, and provide building floorplan data to client computer devices that are authenticated for the tenant entity. The one or more processors of the tenant bounded server system are further configured to receive a set of building floorplan data for a target building included in the platform map data, determine geospatial reference data for the set of building floorplan data by aligning the set of building floorplan data with building outline data of the target building.

Abstract: Innovations in control and use of chroma quantization parameter (“QP”) values that depend on luma QP values. More generally, the innovations relate to control and use of QP values for a secondary color component that depend on QP values for a primary color component. For example, during encoding, an encoder determines a QP index from a primary component QP and secondary component QP offset. The encoder maps the QP index to a secondary component QP, which has an extended range. The encoder outputs at least part of a bitstream including the encoded content. A corresponding decoder receives at least part of a bitstream including encoded content. During decoding, the decoder determines a QP index from a primary component QP and secondary component QP offset, then maps the QP index to a secondary component QP, which has an extended range.

Abstract: Techniques for automatic detection of a presentation surface and generation of an associated data stream include receiving, in connection with a communication session, a first media stream capturing a portion of an environment including a presentation surface; detecting the presence in the first media stream of the presentation surface; detecting usage of the presentation surface during the conferencing session; generating, in response to the detected usage of the presentation surface, a second media stream dedicated to the presentation surface from the first media stream; and transmitting the second media stream dedicated to the presentation surface to one or more recipient devices.

Abstract: A system and method of monitoring a status of a server using a baseboard management controller (BMC) is disclosed. First, the BMC may access a status code generated in Domain 0 of the server through a command line interface to the server and send the status code to an alert management system. Second, the BMC may snoop on a status code generated in the hypervisor of the server through input/output port 80h and send the status code to the alert management system. Third, the BMC may read a status code generated in the hypervisor and written in a Model Specific Register (MSR) and send the status code to the alert management system. The system and method can enable the alert management system to readily recognize the occurrence of an error, including the type of the error which is a software error, and the cause of the error in real time or nearly real time.

Abstract: There is provided a system and method for generating predictions. The predictions are generated using a model configured to associate text with at least one action associated with at least one of a plurality of applications.

Abstract: The description relates to computing devices that employ steerable optics. One example includes a steering mechanism and a base substrate positioned relative to the steering mechanism. The example also includes an optical substrate positioned over the base substrate and an adhesive complex securing the optical substrate relative to the base substrate with multiple different types of adhesives.

Abstract: Systems and methods directed to formulating and suggesting a subject line based on content of an email message are provided. More specifically, content corresponding to one or more portions of an email may be received, where the content may be utilized to determine one or more email topics and at least one intent of the email. An intent of the email may correspond to whether an action is to be performed, whether a question is presented, whether a request for time is present, and/or any other request on the part of the recipient or the sender. Based on the intent of the email and the one or more topics, a subject line suggestion may be formulated and presented to a user.

Abstract: A head-mounted display device is provided, including an at least partially see-through display, an electrical device, and a flexible printed circuit board (FPC) arranged on a surface of a see-through portion of the display. The FPC may include a plurality of electrical traces electrically coupled to the electrical device. Each electrical trace may be separated from at least one other electrical trace by one or more respective see-through gaps. The FPC may further include a transparent material arranged between the electrical traces in each see-through gap.

Abstract: Aspects of the present disclosure relate to techniques for mitigating inadvertent user information collection in telemetry data. In examples, user information is used to evaluate telemetry data associated prior to transmission to a server device. If an instance of user information is identified within the telemetry data, a warning indication is generated. The warning indication may be transmitted to the server device either instead of or in combination with the telemetry data. As a result of the warning indication, the software may be modified to resolve the issue that caused the introduction of the user information into the telemetry data, thereby avoiding future instances of inadvertent data collection. In response to the warning indication, the server may be configured to reject similar telemetry data from other devices, thereby avoiding collecting such data from the other devices.

Abstract: Examples disclosed herein are related to using a machine learning model to generate image data. One example provides a system, comprising one or more processors, and storage comprising instructions executable by the one or more processors to obtain image data comprising an image with unoccluded features, apply a mask to the unoccluded features in the image to form partial observation training data comprising a masked region that obscures at least a portion of the unoccluded features, and train a machine learning model comprising a generator and a discriminator at least in part by generating image data for the masked region and comparing the image data generated for the masked region to the image with unoccluded features.

Abstract: A method, system, and storage device storing a computer program, for generating questions based on provided content, such as, for example, a document having words. The method comprises automatically estimating the probability of interesting phrases in the provided content, and generating a question in natural language based on the estimating. In one example embodiment herein, the estimating includes predicting the interesting phrases as answers, and the estimating is performed by a neural model. The method further comprises conditioning a question generation model based on the interesting phrases predicted in the predicting, the question generation model generating the question. The method also can include training the neural model. In one example, the method further comprises identifying start and end locations of the phrases in the provided content, and the identifying includes performing a dot product attention mechanism parameterizing a probability distribution.

Abstract: A system is described herein. The system includes an agent and a transmitter. The agent is to a capture room specific data signals from peripheral devices within a conference room, operating system data signals, and application data signals, wherein the application data signals are obtained from a conferencing service application and wherein the agent is to package the room specific data signals, the operating system data signals, and the application data signals with system metadata into room health data. The transmitter is to transmit the room health data to a management cloud via a communications broker.

Abstract: In representative embodiments, a digital assistant is extended by service activities. A service activity executes as a service as part of the digital assistant. A service activity specifies a set of user data to be provided when invoked, at least one insight used to invoke the service activity in a proactive way and at least one intent used to invoke the service activity in a reactive way, and executable code that is executed when the service activity is invoked. A development environment provides templates to ease development of service activities. Service activities in development can be deployed alongside production services with access to service activities in development restricted to a set of authorized users.

Abstract: Apparatus and methods are disclosed for using machine learning models with private and public domains. Operations can be applied to transform input to a machine learning model in a private domain that is kept secret or otherwise made unavailable to third parties. In one example of the disclosed technology, a method includes applying a private transform to produce transformed input, providing the transformed input to a machine learning model that was trained using a training set modified by the private transform, and generating inferences with the machine learning model using the transformed input. Examples of suitable transforms that can be employed include matrix multiplication, time or spatial domain to frequency domains, and partitioning a neural network model such that an input and at least one hidden layer form part of the private domain, while the remaining layers form part of the public domain.

Abstract: An audio system including an audio device secured within a receptacle for improving voice communication is disclosed. The audio device includes a housing containing a down-firing speaker that is positioned directly above an acoustic reflector formed by surfaces associated with the bottom of the housing and a corresponding interior of the receptacle. An apex protrudes upward from the center of the receptacle toward the diaphragm of the speaker. The acoustic reflector is characterized by a curved volume extending from the apex to an outer peripheral border of the receptacle. The structural features of the audio device and receptacle are configured to significantly improve the quality of sound produced by the audio system such that it fulfills standardized wideband audio performance requirements in a compact package.

Abstract: A system and method for selecting a target device out of a larger group of candidate devices for rendering a response from a virtual assistant to an end-user is disclosed. The system determines that a same trigger phrase included in an utterance has been received by multiple devices that are in proximity to one another at around the same time. These candidate devices can collect attention data, such as user gaze toward a device, to select the device that was most likely the intended recipient of the utterance. The system is configured to control the virtual assistant to render a response solely via the selected device.