Simultaneous translation of lectures and speeches

With increasing globalization, communication across language and cultural boundaries is becoming an essential requirement of doing business, delivering education, and providing public services. Due to the considerable cost of human translation services, only a small fraction of text documents and an even smaller percentage of spoken encounters, such as international meetings and conferences, are translated, with most resorting to the use of a common language (e.g. English) or not taking place at all. Technology may provide a potentially revolutionary way out if real-time, domain-independent, simultaneous speech translation can be realized. In this paper, we present a simultaneous speech translation system based on statistical recognition and translation technology. We discuss the technology, various system improvements and propose mechanisms for user-friendly delivery of the result. Over extensive component and end-to-end system evaluations and comparisons with human translation performance, we conclude that machines can already deliver comprehensible simultaneous translation output. Moreover, while machine performance is affected by recognition errors (and thus can be improved), human performance is limited by the cognitive challenge of performing the task in real time.     

High Gd Concentration GaGdN Grown at Low Temperatures

GaGdN layers were grown at temperatures below 300°C by radio-frequency plasma-assisted molecular beam epitaxy on sapphire substrates. GaGdN samples with high Gd concentration as high as 12.5% were obtained by lowering the growth temperature. X-ray diffraction results showed no obvious secondary phase, which means that the phase separation can be suppressed by the growth at low temperatures. All samples, including those grown at room temperature, showed ferromagnetic characteristics. Photoluminescence emission was observed, though spectra exhibit broad and sharp luminescence bands related to many kinds of defects. It is suggested that electrons coming from defects, especially, nitrogen vacancy, stabilize ferromagnetism, and that the carrier-induced ferromagnetism occurs in the low-temperature-growth GaGdN.     

Mutual recognition of the results of measurements and tests and formation of a world-wide metrology system

The main prerequisites are considered in a historical aspect and steps ate taken in creating a world-wide metrology system with the aim of providing conditions for the mutual recognition of measurement results. __________ Translated from Izmeritel’naya Tekhnika, No. 11. pp. 59–62, November, 2007.     

The time course of spoken word learning and recognition: Studies with artificial lexicons.

The time course of spoken word recognition depends largely on the frequencies of a word and its competitors, or neighbors (similar-sounding words). However, variability in natural lexicons makes systematic analysis of frequency and neighbor similarity difficult. Artificial lexicons were used to achieve precise control over word frequency and phonological similarity. Eye tracking provided time course measures of lexical activation and competition (during spoken instructions to perform visually guided tasks) both during and after word learning, as a function of word frequency, neighbor type, and neighbor frequency. Apparent shifts from holistic to incremental competitor effects were observed in adults and neural network simulations, suggesting such shifts reflect general properties of learning rather than changes in the nature of lexical representations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Polarity control of GaN epilayers grown on ZnO templates

We report the successful growth of Ga-polar GaN epilayers on O-polar ZnO templates pre-deposited on c-sapphire. Prior to GaN growth, NH₃ is exposed onto the ZnO template. The polarity of the GaN layers is confirmed by etching of the surface and by conversion beam electron diffraction (CBED), while the O-polar ZnO is confirmed by CBED. It is suggested that the NH₃ pre-exposure helps form a Zn₃N₂ layer, which possesses inversion symmetry and inverts the crystal from anion polar to cation polar.     

基于轮廓线匹配的二维重迭物体的识别

本文提出一种基于轮廓线匹配的二维重迭物体的识别算法,并采用一种点模式匹配的快 速算法进行特征点序列的匹配,通过引进形状移位数的概念和利用问题中的各种约束条件,提 高了物体识别速度.     

基于组织协同进化分类算法的遥感图像目标识别

针对遥感图像目标识别问题,提出了一种基于组织协同进化分类算法的识别方法。它没有复杂的运算,训练和识别速度都很快。对实测遥感图像的实验表明,本文方法性能稳定,优于文献[3]和[4]中基于支撑矢量机的方法,识别率均达到了95％以上,且训练时间非常短,不到1秒钟。     

Scopolamine impairs human recognition memory: Data and modeling.

Eight subjects studied a set of complex visual images after administration of 0.4 mg scopolamine. Another 8 subjects performed the same task without drug administration. On a subsequent item recognition test, subjects rated, on a 5-point scale, their confidence that the studied pictures and an equal number of unstudied lures were actually presented. Results showed that scopolamine affected responses to studied items, but not unstudied lures, demonstrating an unambiguous effect of scopolamine on recognition memory. To describe the scopolamine-injected subjects' data, the authors constructed a new model of 2-process recognition that includes the A. P. Yonelinas (1994) model as a limiting case. The model analysis suggests that scopolamine affected both familiarity and recollection. In particular, scopolamine did not affect the frequency with which recollection took place, but rather, affected the amount of recollected information. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A New Method for Target Recognition

1　Introduction　SinceZadeh[1] establishedfuzzysettheoryin 196 5 ,fuzzysetmakesgreatachievementintheoryandapplica tion[2～ 5] .Becausethereexistintrinsicfuzzinessinpat ternrecognition ,usingfuzzysettopatternrecognitionacquiregreatsuccess.Toexplainthedifferentandrela tionbetweenfuzzysetandprobabilitytheory ,Zadeh[6 ]introducedpossibilitytheoryin 1978.Possibilitytheoryhadbeenusedinfuzzyreason ,fuzzycontrolandexpertsystem[7～ 9] .Applingpossibilitytheoryto patternrecognitionisattemptedbyHalland…     

信息融合的飞机识别方法

以信息融合的理论为基础，利用从可见光图像序列和8-12μm长波红外图像序列中提取的信息对不同种类飞机进行识别。采用矩特征并结合BP神经网络的方法，分别在特征级和决策级两个不同层次上实现了信息融合。实验结果表明，通过信息融合进行飞机识别的准确率可达到90％以上。