Two-stage ASGD framework for parallel training of DNN acoustic models using Ethernet.

Zhichao Wang Xingyu Na Xin Li Jielin Pan Yonghong Yan

Published in: ASRU (2015)

Keyphrases

training process
bayesian framework
information retrieval
feature selection
neural network
video sequences
pattern recognition
probabilistic model
multi modal
em algorithm
video retrieval