| Document Date: 2014-10-20 08:22:50
Open Document File Size: 201,10 KBShare Result on Facebook
City Pittsburgh / / Company TTS / Amazon / BP / Neural Networks / Recurrent Neural Networks / / Country United States / Singapore / / / Event Natural Disaster / / Facility Shanghai Jiao Tong University / / IndustryTerm chosen systems / recurrent networks / speech synthesis applications / deep belief network / main stream technology / stochastic gradient descent algorithm / parameter generation algorithms / feed-forward network / research groups / hand-held device / / Organization DNN / US Federal Reserve / LSTM RNN / Shanghai Jiao Tong University / Shanghai / / Person Layer ht Waveform / B. Kingsbury / R. A. Gopinath / P. Nguyen / Backward Layer / H. Lu / O. Watts / V. Vanhoucke / T. Sainath / S. King / / Position original speaker / speaker / model the relationship / native speaker / Forward / / PublishedMedium Machine Learning / The Journal of Machine Learning Research / / RadioStation DNN / / Technology speech recognition / main stream technology / neural network / conventional EM algorithm / BPTT algorithm / Speech parameter generation algorithms / machine learning / stochastic gradient descent algorithm / / URL www.mturk.com/mturk/welcome / http /
SocialTag |
INTERSPEECHTTS Synthesis with Bidirectional LSTM based Recurrent Neural Networks Yuchen Fan 1,2*, Yao Qian2, Fenglong Xie2, Frank K. Soong2 1