2015年4月24日 (五) 06:54的版本

Task To Do

1,  RNN speech recognition (Tied-context-dependent-state and End-to-End)  --Chao Liu/Zhiyuan Tang
2,  Real environment noise cancellation(RNN-DAE: echo or reverberation) --Zhiyong Zhang
3,  Integrate the class information to HCLG fst for speech recognition
4,  Multi-Mode features based VAD --Shi Yin, DONE
5,  DNN based Language identification and Speaker identification
6,  Distant speech recognition(Reverberation, Mutli-microphones) --(Lasso),Xuewei Zhang
7,  Voice conversation --Zhongwei Yao
8,  Unbound activation function(Rectifier/Maxout/Pnorm) go-through searching method. --Zhiyong Zhang
9,  Sparse DNN --Zhiyuan Tang/Chao Liu
10, Neural network visulization --Mian Wang,DONE
11, DNN training GPU parallelization , nnet2 optimization.
12, Hessien-like Monmentum acceleration
13, Correlation based SEONE cluster
14, Dark knowledge --Mengyuan Zhao, Xiangyu Zeng, Zhiyong Zhang, Chao Liu

Technical Report To Write

1, DNN-DAE based noise cancellation -- Xiangyu Zeng / Mengyuan Zhao / Zhiyong Zhang  --DONE
2, Speech Rate DNN speech recognition --Shi Yin/Xiangyu Zeng --DONE
3, CNN+fbank feature combination --Mian Wang /Yiye Lin /Mengyuan Zhao /Shi Yin
4, Uyghur low-resource acoustic model enhancement -- Shi Yin / Mengyuan Zhao / Zhiyong Zhang --DONE
5, Uyghur 20h database release --Kaer /Shi Yin --DONE
6，Dark-Knowledge Transfer
   *: Xiangyu Zeng/ Mengyuan Zhao / Zhiyong Zhang

Paper to Write