Task (computing)

About: Task (computing) is a research topic. Over the lifetime, 9718 publications have been published within this topic receiving 129364 citations.

Low-resource Multi-task Audio Sensing for Mobile and Embedded Devices via Shared Deep Neural Network Representations

Abstract: Continuous audio analysis from embedded and mobile devices is an increasingly important application domain. More and more, appliances like the Amazon Echo, along with smartphones and watches, and even research prototypes seek to perform multiple discriminative tasks simultaneously from ambient audio; for example, monitoring background sound classes (e.g., music or conversation), recognizing certain keywords (‘Hey Siri' or ‘Alexa'), or identifying the user and her emotion from speech. The use of deep learning algorithms typically provides state-of-the-art model performances for such general audio tasks. However, the large computational demands of deep learning models are at odds with the limited processing, energy and memory resources of mobile, embedded and IoT devices. In this paper, we propose and evaluate a novel deep learning modeling and optimization framework that specifically targets this category of embedded audio sensing tasks. Although the supported tasks are simpler than the task of speech recognition, this framework aims at maintaining accuracies in predictions while minimizing the overall processor resource footprint. The proposed model is grounded in multi-task learning principles to train shared deep layers and exploits, as input layer, only statistical summaries of audio filter banks to further lower computations. We find that for embedded audio sensing tasks our framework is able to maintain similar accuracies, which are observed in comparable deep architectures that use single-task learning and typically more complex input layers. Most importantly, on an average, this approach provides almost a 2.1× reduction in runtime, energy, and memory for four separate audio sensing tasks, assuming a variety of task combinations.

Method and system for assigning and distributing work over a computer network

Abstract: The invention is a method and system for aligning a properly skilled external work force with a workflow task for a particular while maintaining control and quality assurance over the work product created for the business by the external workforce. The system enables the assignment and distribution of work via a computer network from a business customer having a particular task that requires a particular skill set, matching that to one or more agents over a computer network, and collecting and assimilating the at least a portion of the final work product for retrieval by the customer.

Microcontroller, data processing system and task switching control method

Abstract: A processor, a task management table, and a scheduler are built in a microcontroller. The processor sequentially runs a plurality of tasks for controlling hardware engines (cores) respectively allocated thereto. The task management table stores task management information which includes state information (ST INFO) representative of the execution state of each task, priority information (PRI INFO) representative of the execution priority of each task, and core identification information (CID INFO) representative of the allocation of the tasks to the cores. The scheduler allows the processor to switch between tasks on the basis of the task management information when a given instruction is decoded or when the execution of any one of the cores is terminated.

Microprogrammed digital data processing system employing tasking at a microinstruction level

Abstract: A microprogrammed data processing system is provided in which each high level instruction is performed by one or more tasks, each task being in turn performed by executing one or more task microinstructions in a microprogrammed manner. Dynamic resource allocation and task synchronization are additionally provided along with a three-stage pipelined architecture so as to provide both multiprogramming and multiprocessing on a microinstruction level.

Combined discrete-event and continuous model simulation and analysis tool

Abstract: The tool comprises the first step of providing a first software component, serving as a timing element, for receiving global synchronization commands as input and issuing global simulation scheduler task dispatch commands as output. A second software component is provided, serving as a global simulation scheduler, for receiving the global simulation scheduler task dispatch commands as input, synchronizing discrete event model and continuous model task dispatch as a function of simulation time, and issuing local simulation scheduler task dispatch commands as output. At least a single third software component is provided, serving as a local simulation scheduler, for receiving the local simulation scheduler task dispatch commands as input and issuing local simulation task execution commands as output. The combination of these steps provides a processing environment wherein the local simulation task execution commands invoke user supplied simulation application tasks in a time synchronized manner.