Abstract: A device includes a memory configured to store a collection of sets of weights, each of the sets of weights representing a respective media segment. The device also includes one or more processors configured to generate data representing the detected first input speech segment and to pass the data representing the detected first input speech segment into a collection of memory units. Each memory unit of the collection of memory units includes a set of weights from the collection of sets of weights. The one or more processors are also configured to generate a first estimate of an associated media segment that represents the detected first input speech segment. The associated media segment corresponds to a first memory unit in the collection of memory units.

Abstract: Embodiments include methods for a user equipment (UE) configured to operate in a radio access network (RAN). Such methods include receiving, from a radio node in the RAN, a new feedback indicator (NFI) that indicates whether acknowledgement signaling previously scheduled for transmission by the UE was received by the radio node. The acknowledgement signaling is associated with one or more first downlink (DL) hybrid ARQ (HARQ) processes. Such methods include, when the NFI indicates the acknowledgement signaling was not received by the RAN node, selecting a first HARQ codebook to be used for encoding the acknowledgement signaling together with further acknowledgement signaling associated with one or more second DL HARQ processes. Such methods include, using the selected first HARQ codebook, encoding the acknowledgement signaling and the further acknowledgement signaling to obtain first encoded information and transmitting the first encoded information to the radio node via an uplink (UL) channel.

Abstract: A memory system includes a memory array having a plurality of memory cells; and a controller coupled to the memory array, the controller configured to: designate a storage mode for a target set of memory cells based on valid data in a source block, wherein the target set of memory cells are configured with a capacity to store up to a maximum number of bits per cell, and the storage mode is for dynamically configuring the target set of memory cells in as cache memory that stores a number of bits less per cell than the corresponding maximum capacity.

Abstract: A device includes one or more processors configured to input one or more segments of an input media stream into a feature extractor. The one or more processors are further configured to pass an output of the feature extractor into an utterance classifier to produce at least one representation of at least one utterance class of a plurality of utterance classes. The one or more processors are further configured to pass the output of the feature extractor and the at least one representation into a segment matcher to produce a media output segment identifier.

Abstract: Systems and techniques are described herein for encoding and/or decoding audio information. For example, a process can process an input audio segment to generate a representation of the input audio segment, and can compare the representation of the input audio segment to representations stored in a memory. The representations represent a plurality of audio segments. The process can determine, based on the comparison, target representation(s) of target audio segment(s) from the representations stored in the memory. The process can determine one or more indices associated with the target audio segment(s). The process can then packetize the one or more indices and transmit the one or more packetized indices (e.g., to a decoder configured to decode the packetized indices).

Abstract: A distributed imaging system for augmented reality devices is disclosed. The system includes a computing module in communication with a plurality of spatially distributed sensing devices. The computing module is configured to process input images from the sensing devices based on performing a local feature matching computation to generate corresponding first output images. The computing module is further configured to process the input images based on performing an optical flow correspondence computation to generate corresponding second output images. The computing module is further configured to computationally combine first and second output images to generate third output images.

Abstract: A display system can include a head-mounted display configured to project light to an eye of a user to display virtual image content at different amounts of divergence and collimation. The display system can include an inward-facing imaging system images the user's eye and processing electronics that are in communication with the inward-facing imaging system and that are configured to obtain an estimate of a center of rotation of the user's eye. The display system may render virtual image content with a render camera positioned at the determined position of the center of rotation of said eye.

Abstract: A multi-phase buck-boost converter includes a first half-bridge circuit, a second half-bridge circuit, a third half-bridge circuit, and a control circuit. The first half-bridge circuit is coupled to a first inductor terminal. The second half-bridge circuit is coupled to a second inductor terminal. The third half-bridge circuit is coupled to a third inductor terminal, a system voltage terminal, and a battery terminal. The control circuit is coupled to the first half-bridge circuit, the second half-bridge circuit, and the third half-bridge circuit. The control circuit is configured to transition the first half-bridge circuit, the second half-bridge circuit, and the third half-bridge circuit from operation in a buck mode to operation in a buck-boost mode based on an off-time of the first half-bridge being less than a particular time.

Abstract: Methods and systems for matching user information between data sets, while preserving data privacy are provided. Some examples relate to matching subsets of users from a first device to subsets of users from a second device, based on a first set of indications and a second set of indications, respectively, to calculate how long it takes for a user to travel to a location of interest, after the user is provided with directed content. A conversion rate may be determined based on how many instances at which a user travels to the location of interest, within a conversion window, after the user is provided with the directed content. The conversion rate may be compared to a baseline conversion rate to determine a change in conversion rate. The change in conversion rate may correspond to an impact of the directed content in causing the user to travel to the location of interest. User data is not shared from the first device to the second device, and vice-versa.

Abstract: This disclosure describes a system, method, and non-transitory computer readable media for an ultrasound probe configured to capture ultrasound images of an examination region. The system includes a first set of one or more sensors coupled to the ultrasound probe and configured to estimate a first positional information associated with the ultrasound probe. The system includes a second set of one or more sensors coupled to the ultrasound probe and configured to capture electromagnetic force (EMF) measurements in the examination region to estimate a second positional information associated with the ultrasound probe. The second positional information is used to calibrate the first set of one or more sensors. The system includes a controller configured to use at least one of (i) the first positional information, or (ii) the second positional information to generate a reconstruction of the examination region based on ultrasound images captured by the ultrasound probe.

Abstract: Systems and methods for a wearable “exo-shell” to improve the gait of elderly people during obstacle avoidance tasks are disclosed. With payload and energy expenditure as a main focus of this design, the present system leverages switchable, passive systems, in combination with lightweight materials that minimize additional metabolic costs, while remaining as “transparent” to the user as possible when inactive.

Abstract: Certain aspects of the present disclosure provide techniques for providing sentiment analysis of content. In order to determine the overall sentiment of content, a request is received by a sentiment analyzer, which then identifies a content identification number and retrieves comments associated with the content identification number. The sentiment analyzer pre-processes the comments, which includes removing all personal identifying information from the comments. The sentiment analyzer sends the pre-processed comments to a natural language processing service, and in turn, receives sentiment indications corresponding to the comments provided. Based on the sentiment scores, the sentiment analyzer generates a sentiment analysis and displays the sentiment analysis in the graphical user interface generated by the sentiment analyzer.

Abstract: Examples described herein relate to offload reliable transport management to a network interface device and store packets to be resent, based on received packet receipt acknowledgements (ACKs), into one or more kernel space queues that are also accessible in user space.

Abstract: A strobel for an article of footwear includes a base layer and an upper layer formed of a polymer material covering a portion of the top surface of the base layer, leaving a portion of the top surface of the base layer exposed about an entire periphery of the base layer. The upper layer is directly bonded to the base layer without any fastening members. A method of forming the strobel is also disclosed.

Abstract: A method includes accessing training images that collectively depict multiple stages of a surgical procedure. The method also includes accessing labels that indicate, for each of the training images, characteristics of one or more surgical tools depicted and a stage of the multiple stages of the surgical procedure depicted. The method also includes training a computational model, using the training images and the labels, to associate runtime images with a stage of the multiple stages based on characteristics of one or more surgical tools that are depicted by the runtime images. Another method includes associating, using a computational model, runtime images with a stage of a surgical procedure based on characteristics of one or more surgical tools depicted by the runtime images and generating output that indicates the stage associated with each of the runtime images.

Abstract: A device may determine that a failure occurred in an emergency call between a public-safety answering point (PSAP) and a first user device associated with an emergency caller, and may identify, based on determining the failure, one of a first option, a second option, or a third option for redirecting an emergency callback to a contact of the emergency caller. The device may cause, when the first option is identified, the emergency callback to be redirected to a second user device associated with the contact of the emergency caller, and may cause, when the first option is identified, a bearer path to be established between the PSAP and the second user device for the emergency callback.

Abstract: A method of generating an output movement layout for a traffic intersection is disclosed which includes receiving intersection geographical data, establishing a center point for the intersection, receiving vehicle global positioning system (GPS) data, establishing a radius of interest for the intersection based on the received vehicle GPS data, establishing entry and exit headings for each vehicle based on the GPS data, generating an angular cluster chart based on the entry and exit headings of each vehicle, and generating an output movement layout for the intersection based on the generated angular cluster chart. Further, a method for green time reallocation at traffic signals is disclosed that is based on the congestion experienced by vehicles identified as following specific movements.

Abstract: Certain aspects of the disclosure provide systems and methods for training an information extraction transformer model architecture directed to pre-training a first multimodal transformer model on an unlabeled dataset, training a second multimodal transformer model on a first labeled dataset to perform a key information extraction task processing the unlabeled dataset with the second multimodal transformer model to generate pseudo-labels for the unlabeled dataset, training the first multimodal transformer model based on a second labeled dataset comprising one or more labels, the pseudo-labels generated, or combinations thereof to generate a third multimodal transformer model, generating updated pseudo-labels based on label completion predictions from the third multimodal transformer model, and training the third multimodal transformer model using a noise-aware loss function and the updated pseudo-labels to generate an updated third multimodal transformer model.

Abstract: Aspects of the subject technology provide for suggested home screens and/or lock screens of an electronic device. The suggested home screens and/or lock screens can include suggested backgrounds, user interface elements, colors, styles, photos, and the like, and can be generated and/or suggested for a user of an electronic device based on activity history information for the user and/or the electronic device. In one or more implementations, the home screens and/or lock screens may be generated and/or suggested specifically for various operational modes of the electronic device.

Abstract: Systems and methods are described for generating dynamic web clippings. A web clipping service can obtain from a client computing device metadata for a dynamic web clipping, which includes an identifier of a web document and state information of a web browser that has rendered the web document on the client computing device, the state information enabling replication of a browsing session of the client computing device. The web clipping service can use the metadata to recreate the browsing session of the client device on the web clipping service, and capture an image of the web document as rendered within he recreated browsing session. The web clipping service can then provide the image of the web document as the dynamic web clipping.