This paper introduces a large-scale spontaneous speech corpus of Korean, named KsponSpeech. This corpus contains 969 h of general open-domain dialog utterances, spoken by about 2000 native Korean speakers in a clean environment. All data were constructed by recording the dialogue of two people freely conversing on a variety of topics and manually transcribing the utterances. The transcription provides a dual transcription consisting of orthography and pronunciation, and disfluency tags for spontaneity of speech, such as filler words, repeated words, and word fragments. This paper also presents the baseline performance of an end-to-end speech recognition model trained with KsponSpeech. In addition, we investigated the performance of standard end-to-end architectures and the number of sub-word units suitable for Korean. We investigated issues that should be considered in spontaneous speech recognition in Korean. KsponSpeech is publicly available on an open data hub site of the Korea government.
A case of tick bite was found in the inguinal region of a 74-year-old Korean woman. She was attacked by the tick while working in her vegetable garden in the vicinity of mountain located in Suncheon City, the southern coastal area of the Korean Peninsula. On admission she complained of mild discomfort and itching around the bite area. The causative tick was 23 mm long and had slender pedipalps. The scutum was quite ornate and had eyes at the edge. The genital aperture was located anterior to the level of the coxa II. The spiracular plate was comma-shaped and the anus was surrounded posteriorly by the anal groove. The coxa I had subequal 2 spurs; the external one slightly larger. The spur of coxa IV was slightly longer than those of coxae II and III. The tarsus IV had 2 distinct subapical ventral spurs. It was identified as the fully engorged adult female of Amblyomma testudinarium. This is the first human case of Amblyomma bite in Korea.
In this paper, the robustness of the artificial neural networks to noise is demonstrated with a multilayer perceptron, and the reason of robustness is due to the statistical orthogonality among hidden nodes and its hierarchical information extraction capability. Also, the misclassification probability of a well‐trained multilayer perceptron is derived without any linear approximations when the inputs are contaminated with random noises. The misclassification probability for a noisy pattern is shown to be a function of the input pattern, noise variances, the weight matrices, and the nonlinear transformations. The result is verified with a handwritten digit recognition problem, which shows better result than that using linear approximations.
scite is a Brooklyn-based organization 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.