Abstract: Prompt discovery is performed for identifying prompts to natural language processing machine learning models. A request to determine a prompt for a natural language processing task performed by a pre-trained natural language processing machine learning model may be received. A task classification for the natural language processing task may be determined and candidate prompts for the natural language processing prompt task collection selected. Respective prompt results for the candidate prompts are evaluated to generate a prompt recommendation for the natural language processing task.

Abstract: Attachment of a pluggable module to an externally-accessible slot of a base unit of a server is detected. The module is configured to execute a first network function of a radio-based communication network. In response to a determination that the module satisfies a security criterion, a second network function is launched. The second network function performs one or more computations on output of the first network function. The output of the first network function is generated at the module in response to a message from a user equipment device of a radio-based communication network.

Abstract: Time and value ordering may be applied for items stored in data backups. A change log that persists changes to a data set may be updated with changes and used to update an in-memory table for the data set, which describes changes to items up to a current time. An event may be detected to seal the in-memory table from subsequent updates and a persistent data object that orders the items in the in-memory according to both keys of the respective items and the respective time values of the items, as stored in the change log, may be generated and stored as part of a backup for the data set.

Abstract: The A/V recording and communication devices used in a security system may be configured to detect motion, and to track and report that motion to a building resident. Some residents may wish to customize the tracking and reporting of motion based upon where and when the motion is detected. For example, a user may wish to be alerted when cars drive down her residential street at night, but does not wish to be alerted when cars drive down the street during the day. The present embodiments disclose several tools and methods by which a building resident can configure and manage the tracking and reporting of detected motion. To do so, the resident should configure two or more motion zones within the field of view captured by a camera, together with a set of time windows during which she wishes to track or be notified of detected motion.

Abstract: A voice controlled assistant has a housing to hold one or more microphones, one or more speakers, and various computing components. The housing has an elongated cylindrical body extending along a center axis between a base end and a top end. The microphone(s) are mounted in the top end and the speaker(s) are mounted proximal to the base end. The microphone(s) and speaker(s) are coaxially aligned along the center axis. The speaker(s) are oriented to output sound directionally toward the base end and opposite to the microphone(s) in the top end. The sound may then be redirected in a radial outward direction from the center axis at the base end so that the sound is output symmetric to, and equidistance from, the microphone(s).

Abstract: Exemplary embodiments relate to adapting a generic language model during runtime using domain-specific language model data. The system performs an audio frame-level analysis, to determine if the utterance corresponds to a particular domain and whether the ASR hypothesis needs to be rescored. The system processes, using a trained classifier, the ASR hypothesis (a partial hypothesis) generated for the audio data processed so far. The system determines whether to rescore the hypothesis after every few audio frames (representing a word in the utterance) are processed by the speech recognition system.

Abstract: Customers of a shared-resource environment can provision resources in a fine-grained manner that meets specific performance requirements. A customer can provision a data volume with a committed rate of Input/Output Operations Per Second (IOPS) and pay only for that commitment (plus any overage), and the amount of storage requested. The customer will then at any time be able to complete at least the committed rate of IOPS. If the customer generates submissions at a rate that exceeds the committed rate, the resource can still process at the higher rate when the system is not under pressure. Even under pressure, the system will deliver at least the committed rate. Multiple customers can be provisioned on the same resource, and more than one customer can have a committed rate on that resource. Customers without committed or guaranteed rates can utilize the uncommitted portion, or committed portions that are not being used.

Abstract: Journaling for files in a distributed file system may be dynamically changed. A write request for a file in a distributed file system may be received. An evaluation of the file may be performed with respect to a journaling type change criteria. If the evaluating indicates a change to journaling type, then the write request may be performed according to a different type of journaling than was applied to prior write request. If the evaluating does not indicate a change, then a current journaling type may be applied to perform the write request.

Abstract: According to aspects of the present disclosure, user-defined settings may be assigned to a dummy device in response to a first request. A device identifier of a particular electronic device may later be registered with the dummy device in response to a second request. Based on receiving data indicating an installation procedure of the particular electronic device and based on the device identifier being registered with the dummy device, user-defined settings may be transmitted to cause configuration of at least some of the user-defined settings on the particular electronic device. In some cases, the data indicating an installation procedure corresponds to a beacon from the particular electronic device (e.g., a smart home device), which includes the device identifier. Further, the user-defined settings may be provided to the particular electronic device by another electronic device (e.g., another smart home device).

Abstract: Burst performance of a database query may be determined according to a size of the database query. A query to a database may be received. A size may be determined for the query. If the size is less than a size threshold assigned to a first query engine, then the query may be performed at the first query engine. If the size is greater than or equal to the size threshold assigned to the first query engine, then the query may be performed at a second query engine.

Abstract: An interface for requesting, and technique for generation of, a backup of a past state of a database table are provided. Changes made to a database table are accumulated, in durable storage, and snapshots of partitions of the table are obtained. The accumulated changes and the successive partition snapshots are used to generate a past state of the database at any point in time across a continuum between successive snapshots. Although each partition of the table may have a snapshot that was generated at a time different from when other partition snapshots were generated, changes from respective change logs may be selectively log-applied to distinct partitions of a table to generate backup in the past of the entire table at common point-in-time across partitions.

Abstract: Pre-fix matching may constrain the generation of next token predictions. Input text to perform a next token prediction may be received. Multiple tokens may be determined from the input text, including a partial token. From possible tokens, one or more matching possible tokens with the partial token may be identified. Next token predictions may then be filtered using the identified possible tokens in order to ensure that the partial token is matched.

Abstract: Approaches are disclosed for providing path diversity in a data transmission network. A primary transmission path can be selected through a network, such as a backbone network, based on factors such as cost of transmission. At least one waypoint can be selected that is to be included in a secondary transmission path. The waypoint(s) can be selected such that the secondary transmission path will have few, if any, network components in common with the primary transmission path, providing significant path diversity. The waypoint(s) can be selected based on a cost ratio or other such factor. In the event of a failure of transmission of a data packet over one of the transmission paths, a second transmission attempt can be performed using the same path or the other transmission path, or both.

Abstract: Systems and methods are described for a multi-tenant mode of a serverless code execution system. For instance, a method may include maintaining a set of execution environments, wherein each execution environment is associated with a serverless function, wherein the serverless function is associated with a software as a service (SaaS) provider that is a tenant of a cloud services provider, wherein the SaaS provider provides services to sub-tenants, wherein the set of execution environments are partitioned based on sub-tenants of the SaaS provider; receiving a call to execute a serverless function, wherein the call includes a serverless function identifier and a sub-tenant identifier; identifying a sub-tenant-specific execution environment of the set of execution environments that is associated with the sub-tenant; and in response to identifying the tenant-specific execution environment, invoking the serverless function on the sub-tenant-specific execution environment.

Abstract: Techniques for creating groups of devices for controlling these groups with voice commands are described herein. For instance, an environment may include an array of secondary devices (or “smart appliances”, or simply “devices”) that are configured to perform an array of operations. Users may request to create different groups of these devices, such that the users may control entire groups at a single time with individual voice commands.

Abstract: Techniques for training and using a machine learning model for estimation of reverberation in a multi-task learning framework are described. According to some embodiments, the multi-task learning framework improves the performance of the machine learning model by estimating the amount of reverberation present in an input audio recording as a secondary task to the primary task of generating a clean speech portion of the input audio recording. In one embodiment, a model architecture is selected that takes a noisy reverberant recording as an input and outputs an estimate of a clean (e.g., de-reverberated) signal, an estimate of noise (e.g., background noise), and an estimate of the reverb only portion, with the secondary task of estimating the reverb only portion acting as a regularizer that improves the machine learning model's performance in enhancing the reverberant (e.g., and noisy) input speech.

Abstract: Techniques and apparatus describe a robotic workcell for performing a robotic stowing operation. The robotic workcell includes a mechanism system, which includes a first robotic gantry, a second robotic gantry, and a controller. The first robotic gantry includes a first mount and a robotic end effector(s) attached to the first mount. The second robotic gantry includes a second mount and at least one end effector attached to the second mount. The controller controls movement of the first mount, the second mount, and the robotic end effector(s), based in part on an analysis of an environment with one or more perception sensors.

Abstract: Disclosed are systems and methods that detect segments of video, such as HDR video, that include content, such as edges and details in dark scenes, that cannot be presented on some displays. Output models for different display types, such as edge-lit LCD, backlit LCD, etc., may be created and used to process video with respect to those different display types to determine if segments of the video cannot be presented on the display type at the pixel brightness values indicated in the video. In some implementations, HDR video may also or alternatively be compared to SDR video to determine segments of the video that are of interest, especially in low light scenes.