End-to-end Speech Translation (ST) models have several advantages such as lower latency, smaller model size, and less error compounding over conventional pipelines that combine Automatic Speech Recognition (ASR) and text Machine Translation (MT) models. However, collecting large amounts of parallel data for ST task is more difficult compared to the ASR and MT tasks. Previous studies have proposed the use of transfer learning approaches to overcome the above difficulty. These approaches benefit from weakly supervised training data, such as ASR speech-to-transcript or MT textto-text translation pairs. However, the parameters in these models are updated independently of each task, which may lead to sub-optimal solutions. In this work, we adopt a metalearning algorithm to train a modality agnostic multi-task model that transfers knowledge from source tasks=ASR+MT to target task=ST where ST task severely lacks data. In the meta-learning phase, the parameters of the model are exposed to vast amounts of speech transcripts (e.g., English ASR) and text translations (e.g., English-German MT). During this phase, parameters are updated in such a way to understand speech, text representations, the relation between them, as well as act as a good initialization point for the target ST task. We evaluate the proposed meta-learning approach for ST tasks on English-German (En-De) and English-French (En-Fr) language pairs from the Multilingual Speech Translation Corpus (MuST-C). Our method outperforms the previous transfer learning approaches and sets new state-of-the-art results for En-De and En-Fr ST tasks by obtaining 9.18, and 11.76 BLEU point improvements, respectively.
In July, 2015, the Korean national assembly passed 'Act on Capital Markets and Financial Investments,' and therefore, it was expected that the crowd funding would be activated owing to a variety of fundraisings and investments. Hence, for the success of the crowd funding, this paper tried to identify the factors affecting the funding. In this study we analyzed the core variables of the Unified Theory of Acceptance and Use of Technology(UTAUT) and their perceived risks on the crowd funding participants' intentions as well as the mediating effects of the attitudes; the core variables of UTAUT were performance expectancy, perceived risk, facilitating conditions, social influence, and the like. As a result, it was found that such facilitating conditions as performance expectancy and social influence would affect crowd funding participants' intention positively, but that effort expectancy and perceived risk would not significantly affect their intention. On the other hand, as a result of testing the mediating effects of the attitudes, it was found that performance expectancy and social influence would have significant mediating effects on participants' intention.
Geographic routing has been addressed in many literatures of ad hoc sensor networks due to its efficiency and scalability. Void areas (holes) bring Geographic routing some problems such as data congestion and excessive energy consumption of hole boundary nodes. Holes are hardly avoided in wireless sensor networks due to various actual geographical environments, e.g., puddles, buildings or obstacles, or uneven energy consumption, even physical destruction. To bypass a hole, most existing geographic routing protocols tend to route data packets along the boundary of the hole by perimeter routing scheme. This scheme, on one hand, consumes more energy of the nodes on the boundary of the hole, thus possibly enlarging the hole, we call this hole diffusion problem; on the other hand, it may incur data congestion if multiple communication sessions are bypassing the hole simultaneously. In this paper, we propose Efficient Hole Detour Scheme to solve the hole problems faced by geographic routing in wireless sensor networks. Simulation results show that the proposed protocol is superior to other protocols in terms of packet deliver ratio, control overhead, average delivery delay, and energy consumption.
scite is a Brooklyn-based startup that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.