Abstract: Provided is a method for implementing dynamic multi-streaming. The method includes: generating a grouping instruction based on a first user operation, the first user operation being related to a grouping requirement of a user; grouping audio devices in a target area based on the grouping instruction to obtain device grouping information; generating a playback instruction based on the device grouping information and a second user operation, the second user operation being related to a playback requirement of the user, and the playback requirement including a target audio source; and controlling a target device to play the target audio source based on the playback instruction.

Abstract: A method, a system for multi-stream multi-zone interaction, and a non-transitory computer-readable storage medium are provided. The method includes: acquiring an audio source service item, a floor plan of a target region, and a device layout diagram, wherein the audio source service item includes an audio source and an audio source type; determining a zone to be configured corresponding to a playback device based on the floor plan and the device layout diagram; in response to an application operation on the audio source service item, determining at least one of a paired zone corresponding to the audio source service item or an application mode of the audio source service item in the paired zone; wherein the paired zone is the zone to be configured that has a pairing relationship with the audio source service item, and the application mode includes direct application or a mixing operation.

Abstract: The present disclosure relates to a method, a system, and a storage medium for interaction based on natural language, wherein the method is executed by a processor and includes: determining target input information of a user based on initial input information of the user and tag information of an operable object, wherein the tag information reflects a feature of the operable object; determining a target operation based on at least one of the initial input information or the target input information, wherein the target operation includes at least one of a display screen operation, an input correction operation, and an intelligent customer service interaction; and generating an operation feedback based on at least one of the initial input information or the target input information, wherein a feedback interface of the operation feedback includes a user modification window.

Abstract: A tissue sample sectioning device including a supporting matrix having a top surface, configured to accept a tissue sample thereon, a micro-scaffold comprising a top, a bottom, and a plurality of micro-wells, wherein each micro-well is defined by at least one or more walls that extend between the top and the bottom of the micro-scaffold, and each of the one or more walls has a cutting edge at the bottom of the micro-scaffold, and a press configured to apply a force onto the micro-scaffold, such that when a force is applied, the cutting edge cuts the tissue sample, and the tissue sample is sectioned into the plurality of micro-wells. The disclosure also provides a method of sectioning a tissue sample using the supporting matrix and the micro-scaffold by applying a force.

Abstract: Command stream stitching for hardware acceleration includes generating, by a host processor, a stitched block representing a plurality of commands for a hardware accelerator. The host processor generates a stitched command from the plurality of commands. The stitched command references the stitched block. The hardware accelerator executes the stitched block in response to invoking the stitched command. The hardware accelerator generates a single notification directed to the host processor for the stitched command.

Abstract: The present invention provides phenylglyoxal-based alkyne (PGA) chemical tags exhibiting high specificity towards protein citrullination sites and other biomolecules containing similarly reactive functional groups. The PGA tags are able to bind to or derivatize biomolecules, such as polypeptides having one or more post-translational modifications (PTMs), such as citrullination. In particular, the PGA tags of the present invention have superior reactivity and selectivity towards ureido groups, and allow for the analysis of biomolecules containing ureido groups facilitated by click chemistry and mass spectrometry (MS) techniques and methods for qualitative and quantitative analysis of biological and clinical samples.

Abstract: A method and system for determining a spatiotemporal fractionation radiotherapy planning, belonging to the technical field of medical equipment which involves acquiring pathological images; identifying the acquired pathological images to determine the region of interest in the pathological images as the radiotherapy region; arranging the irradiation field direction according to the shape of the radiotherapy region, setting the optimization conditions of the radiotherapy planning according to clinical prescription requirements, and planning the radiotherapy path; evaluating the plan quality based on the planned optimization model for the planned radiotherapy path, and resetting the optimization conditions for unsuitable radiotherapy plans until the plan quality meets the requirements; outputting the radiotherapy path that meets the plan quality requirements as the final spatiotemporal fractionation radiotherapy plan.

Abstract: The present disclosure provides systems and methods for determining priority discrepancies between audio data and records data. In one example, the system includes an electronic computing device. The electronic computing device compares audio data associated with a public safety incident to records data associated with the public safety incident, determines discrepancies between the audio data and the records data, and determines an organization and a type associated with the public safety incident. The electronic computing device also determines, based on the organization and the type, priority discrepancies included in the discrepancies and generates a geographical map that includes the priority discrepancies and that is configured for display on a display device.

Abstract: Embodiments herein describe partitioning an acceleration device based on the needs of each user application executing in a host. In one embodiment, a flexible queue provisioning method allows the acceleration device to be dynamically partitioned by pushing the configuration through a control command queue to the device by management software running in a trusted zone. The new configuration is parsed and verified by trusted firmware, which, then, creates isolated IO command queues on the acceleration device. These IO command queues can be directly mapped to a user application, VM, or other PCIe devices. In one embodiment, each IO command queue exposes only the compute resource assigned by the trusted firmware in the acceleration device.

Abstract: The present invention provides improved cleavable biotin-containing isobaric tags for quantitative mass spectrometry proteomics, particularly useful in quantification of post-translational modifications (PTMs) and assessing cysteine containing proteins. The isobaric chemical tags, DiLeu-Biotin-Azide (DBA), consist of three “modules”: (1) a DiLeu reporter group for relative quantification using MS/MS; (2) a biotin moiety for selective enrichment via streptavidin beads; and (3) an azide functional group to enable biorthogonal click chemistry. The DBA tags could be used in a high-throughput quantitative pan-PTM analysis platform.

Abstract: A method for training a streaming automatic speech recognition student model includes receiving a plurality of unlabeled student training utterances. The method also includes, for each unlabeled student training utterance, generating a transcription corresponding to the respective unlabeled student training utterance using a plurality of non-streaming automated speech recognition (ASR) teacher models. The method further includes distilling a streaming ASR student model from the plurality of non-streaming ASR teacher models by training the streaming ASR student model using the plurality of unlabeled student training utterances paired with the corresponding transcriptions generated by the plurality of non-streaming ASR teacher models.

Abstract: A method for training a streaming automatic speech recognition student model includes receiving a plurality of unlabeled student training utterances. The method also includes, for each unlabeled student training utterance, generating a transcription corresponding to the respective unlabeled student training utterance using a plurality of non-streaming automated speech recognition (ASR) teacher models. The method further includes distilling a streaming ASR student model from the plurality of non-streaming ASR teacher models by training the streaming ASR student model using the plurality of unlabeled student training utterances paired with the corresponding transcriptions generated by the plurality of non-streaming ASR teacher models.

Abstract: A fiber optic module for installation within patch panel, the fiber optic module comprising a first outer wall with a first rail profile on its outside surface, a second outer wall with a second rail profile on its outside surface, and at least one optical connector. The first outer wall and the second outer wall run parallel to each other. The at least one optical connector is arranged between the first outer wall and the second outer wall. The first rail profile and the second rail profile are complementary to each other such that the fiber optic module may be coupled to an adjacent fiber optic module, wherein the first rail profile of the fiber optic module may be coupled to the second rail profile of the adjacent fiber optic module.

Abstract: Embodiments herein describe partitioning an acceleration device based on the needs of each user application executing in a host. In one embodiment, a flexible queue provisioning method allows the acceleration device to be dynamically partitioned by pushing the configuration through a control command queue to the device by management software running in a trusted zone. The new configuration is parsed and verified by trusted firmware, which, then, creates isolated IO command queues on the acceleration device. These IO command queues can be directly mapped to a user application, VM, or other PCIe devices. In one embodiment, each IO command queue exposes only the compute resource assigned by the trusted firmware in the acceleration device.

Abstract: Techniques for generating a user interface (UI) for a public safety user are provided. An event trigger may be received at a computing device associated with a public safety user. The event trigger may be associated with a public safety event. Historical interaction with at least one UI object by the public safety user when handling previous event triggers of the same type as the received event trigger may be determined. The historical interaction may be retrieved from a historical UI interaction data store. UI objects that the public safety user may interact with to handle the received event trigger may be identified based on the determined historical interaction. A UI may be generated to handle the received event trigger, the UI to handle the received event trigger including the identified UI objects.

Abstract: A method for training a streaming automatic speech recognition student model includes receiving a plurality of unlabeled student training utterances. The method also includes, for each unlabeled student training utterance, generating a transcription corresponding to the respective unlabeled student training utterance using a plurality of non-streaming automated speech recognition (ASR) teacher models. The method further includes distilling a streaming ASR student model from the plurality of non-streaming ASR teacher models by training the streaming ASR student model using the plurality of unlabeled student training utterances paired with the corresponding transcriptions generated by the plurality of non-streaming ASR teacher models.

Abstract: The present disclosure provides systems and methods for determining priority discrepancies between audio data and records data. In one example, the system includes an electronic computing device. The electronic computing device compares audio data associated with a public safety incident to records data associated with the public safety incident, determines discrepancies between the audio data and the records data, and determines an organization and a type associated with the public safety incident. The electronic computing device also determines, based on the organization and the type, priority discrepancies included in the discrepancies and generates a geographical map that includes the priority discrepancies and that is configured for display on a display device.

Abstract: Methods, systems, and apparatus, including computer programs stored on a computer-readable storage medium, for transliteration for speech recognition training and scoring. In some implementations, language examples are accessed, some of which include words in a first script and words in one or more other scripts. At least portions of some of the language examples are transliterated to the first script to generate a training data set. A language model is generated based on occurrences of the different sequences of words in the training data set in the first script. The language model is used to perform speech recognition for an utterance.

Abstract: A method for handling an overload state in a public safety answering point (PSAP) is provided. While an overload state is not detected at a call routing computing device received emergency calls are queued for a human operator. Responsive to detecting an overload state, first identifying those calls initiated from mobile devices capable of communicating via text message and transmitting a text message requesting callers to respond via text identifying the matter of the emergency call. The responsive text messages are classified as one of a verified emergency call and a verified non-emergency call to be returned when the overload is no longer detected. For each verified emergency call the call is queued for answer by a human operator. For each verified non-emergency call, a follow up text message is transmitted indicating that the verified non-emergency call will be returned at a later time.