http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
육동석,임단,유인철 한국음향학회 2020 韓國音響學會誌 Vol.39 No.5
Sequence-to-sequence deep neural networks with attention mechanisms have shown superior performance across various domains, where the sizes of the input and the output sequences may differ. However, if the input sequences are much longer than the output sequences, and the characteristic of the input sequence changes within a single output token, the conventional attention mechanisms are inappropriate, because only a single context vector is used for each output token. In this paper, we propose a double-attention mechanism to handle this problem by using two context vectors that cover the left and the right parts of the input focus separately. The effectiveness of the proposed method is evaluated using speech recognition experiments on the TIMIT corpus.
육동석,서형진,고봉구,유인철,Yook, Dongsuk,Seo, HyungJin,Ko, Bonggu,Yoo, In-Chul 한국음향학회 2022 韓國音響學會誌 Vol.41 No.3
Recently, Generative Adversarial Networks (GAN) and Variational AutoEncoders (VAE) have been applied to voice conversion that can make use of non-parallel training data. Especially, Conditional Cycle-Consistent Generative Adversarial Networks (CC-GAN) and Cycle-Consistent Variational AutoEncoders (CycleVAE) show promising results in many-to-many voice conversion among multiple speakers. However, the number of speakers has been relatively small in the conventional voice conversion studies using the CC-GANs and the CycleVAEs. In this paper, we extend the number of speakers to 100, and analyze the performances of the many-to-many voice conversion methods experimentally. It has been found through the experiments that the CC-GAN shows 4.5 % less Mel-Cepstral Distortion (MCD) for a small number of speakers, whereas the CycleVAE shows 12.7 % less MCD in a limited training time for a large number of speakers.
육동석,이효원,유인철 한국음향학회 2020 韓國音響學會誌 Vol.39 No.6
Since a large amount of training data is typically needed to train Deep Neural Networks (DNNs), a parallel training approach is required to train the DNNs. The Stochastic Gradient Descent (SGD) algorithm is one of the most widely used methods to train the DNNs. However, since the SGD is an inherently sequential process, it requires some sort of approximation schemes to parallelize the SGD algorithm. In this paper, we review various efforts on parallelizing the SGD algorithm, and analyze the computational overhead, communication overhead, and the effects of the approximations. 심층 신경망(Deep Neural Network, DNN) 모델을 대량의 학습 데이터로 학습시키기 위해서는 많은 시간이소요되기 때문에 병렬 학습 방법이 필요하다. DNN의 학습에는 일반적으로 Stochastic Gradient Descent(SGD) 방법이 사용되는데, SGD는 근본적으로 순차적인 처리가 필요하므로 병렬화하기 위해서는 다양한 근사(approximation) 방법을 적용하게 된다. 본 논문에서는 기존의 DNN 병렬 학습 알고리즘들을 소개하고 연산량, 통신량, 근사 방법 등을분석한다.
동석호 대한소화기내시경학회 2012 Clinical Endoscopy Vol.45 No.3
The pancreatobiliary organ is composed of one of the most complicated structures and complex physiological functions among other digestive organs in our body. This is why endoscopic procedure in pancreaticobiliary system requires rather complicated techniques. In International Digestive Endoscopy Network (IDEN) 2012, many interesting pancreatobiliay endoscopy related topics were presented. Basic procedures like endoscopic papillary balloon dilation (EPBD), advanced techniques like endoscopic necrosectomy, prevention and management of post-ERCP pancreatitis, and spyglass system are reviewed in this highlight summary.