Jennifer M. Schopf, Ann Chervenak, Ian Foster, Dan Fraser, Dan Gunter, Nick LeRoy, Brian Tierney 1 Computation Institute, University of Chicago and Argonne National Laboratory 2 Mathematics and Computer Science Division, Argonne National Laboratory 3 Information Sciences Institute, University of Southern California 4 Department of Computer Science, University of Chicago 5 Lawrence Berkeley Nati...

We explore the use of compression methods to improve the middleware-based exchange of information in interactive or collaborative distributed applications. In such applications, good compression factors must be accompanied by compression speeds suitable for the data transfer rates sustainable across network links. Our approach combines methods that continuously monitor current network and proce...

Several data collection algorithms, which are based on the combination of using mobile sinks and multiple-hop forwarding, have been proposed to prolong network lifetime wireless sensor networks. However, most approaches treat point selection touring path planning as two independent problems, leads a sub-optimal solution for collection. This article an ant colony optimization end-to-end strategy...

We present a new end-to-end architecture for automatic speech recognition (ASR) that can be trained using symbolic input in addition to the traditional acoustic input. This architecture utilizes two separate encoders: one for acoustic input and another for symbolic input, both sharing the attention and decoder parameters. We call this architecture a multi-modal data augmentation network (MMDA),...

The paper introduces an approach to bridging the gap between diverse data sources and software for visualization and data analysis. The possibility of XML-based languages for data structure description and the introduction of Grid services allows the possibility of partitioning the problem to straightforward components. The work takes place within gViz, a project within the UK escience Core Pro...

Attention-based encoder-decoder (AED) models have achieved promising performance in speech recognition. However, because of the end-to-end training, an AED model is usually trained with speech-text paired data. It challenging to incorporate external text-only data into models. Another issue that it does not use right context a text token while predicting token. To alleviate above two issues, we...

We present a method for introducing text encoder into pre-trained end-to-end speech translation systems. It enhances the ability of adapting one modality (i.e., source-language speech) to another text). Thus, model can learn from both unlabeled and labeled data, especially when data is abundant. Beyond this, we denoising build robust that deal with normal noisy data. Our system sets new state-o...