In this study, the effect of the friction stir welding (FSW) was compared with that of the gas tungsten arc welding (GTAW) on the microstructure and microhardness of Cu-Ni alloy weldment. The weldment of 10 mm thickness was fabricated by FSW and GTAW, respectively. Both weldments were compared with each other by optical microstructure, microhardness test and grain size measurement. Results of this study suggest that the microhardness decreased from the base metal (BM) to the heat affected zone (HAZ) and increased at fusion zone (FZ) of GTAW and stir zone (SZ) of FSW. the minimum Hv value of both weldment was obtained at HAZ, respectively, which represents the softening zone, whereas Hv value of FSW weldment was little higher than that of GTAW weldment. These phenomena can be explained by the grain size difference between HAZs of each weldment. Grain size was increased at the HAZ during FSW and GTAW. Because FSW is a solid-state joining process obtaining the lower heat-input generated by rotating shoulder than heat generated in the arc of GTAW.

A5J32-T4 and A5052-H32 dissimilar aluminum alloy plates with thickness of 1.6 and 1.5 mm were welded by friction stir lap welding (FSLW). The FSLW were studied using different probe length tool and various welding conditions which is rotation speed of 1000, 1500 rpm and welding speed of 100 to 600 mm/min and material arrangement, respectively. The effects of plunge depth of tool and welding conditions on tensile properties and weld nugget formation. The results showed that three type nugget shapes such as hooking, void, sound have been observed with revolutionary pitch. This plunge depth and material arrangement were found to effect on the void and hooking for- mation, which in turn significantly influenced the mechanical properties. The maximum joint efficiency of the FSLWed plates was about 90% compared to base metal, A5052-H32 when the A5052-H32 was positioned upper plate and plunge depth was positioned at near interface between upper and lower plates.

This study aims at unveiling phonetic cue weighting of English word recognition by Korean learners of English using a web-based online perception experiment. The longer duration of a high front tense vowel than that of a lax counterpart is used as a cue by native English speakers and presumably as a cue by Korean learners of English. The longer duration of a vowel before voiced stops than that of an equivalent before voiceless stops serves as a cue by native Speakers of English and also probably by Korean learners of English. A web-based online experiment has been designed to examine how Korean learners of English cope with a mixture of fine grained phonetic cues. Results of the web-based online experiment reveal that duration is primarily and robustly used to distinguish a tense vowel before voiced stops and a lax vowel before voiceless stops. At the same time, spectral cues are also put to use, though to a lesser extent, to help make a distinction from a tense vowel before voiceless stops and a lax vowel before voiced stops. The study implies that learners of English try to utilize a complex mixture of acoustic cues to distinguish words.

Yong-hun Lee, JeeHee Yu, and Tae-Jin Yoon. 2017. Predicting the Occurrence of the English Modals Can and May Using Deep Neural Networks. Studies in Modern Grammar 96, 167-189. This paper tries to provide a computational modeling of language processing using deep neural networks. For this purpose, the corpus data in the ICE-USA was used. After all the sentences with can and may were encoded with eighteen linguistic factors, the annotated data were fed into the deep neural networks (DNN). The DNN was constructed with three layers, and each layer contained seventeen nodes. After the DNN was constructed, the learning process was performed with a training set. Then, the performance was measured with a test set. The processes were repeated one hundred times, and it was observed that the DNN had the classification accuracy of 91.5%. The results are promising in that reliable methods can be used in automatically classifying the frequently used modal auxiliary on the basis of the deep learning system.

Statistical rhythmic metrics are applied on a Buckeye corpus of spontaneous interview speech in order to investigate the extent of inter-speaker rhythm variability. Tests are made on speech produced by speakers who share the same regional dialect in North America. The choice is made due to the unique characteristic of the Buckeye corpus in that the speech dataset is obtained from the speakers who have been raised in the same region and hence who share the same dialect with each other. Statistical measures of rhythm metrics are obtained from the subset of the corpus. The results of clustering analysis show that the rhythmic measures that capture the least dialectal variance is the normalized pair-wise variability indices calculated based on the respective adjacent consonantal and vocalic durations. The finding implies that these statistical measures of rhythm can be used in capturing the dialectal similarities on spontaneous speech.

The paper examines the relative contribution that prominence effect and frequency effect make on the phonetic substances in a large-scale corpus of reading-style Standard Korean. Japanese has been reported to have a sentence-final shortening effect in reading style database, probably due to frequency effect with attenuated effect of prosodic prominence. To examine whether the Korean reading style has a similar effect to the Japanese sentence-final shortening, phonetic characteristics of declarative sentence-ending ‘ta’ are examined based on the speech of 10 male and 10 female Korean speakers drawn from a large-scale speech corpus. The result showed that the declarative-ending ‘ta’ is marked by higher F1 as well as longer duration and lower F0 than medial ‘ta,' which served as a reference. The paper ends with the discussion of phonological consequences that these phonetic prominence effects may have in a number of languages.