2015年1月14日 (三) 06:39的版本

Task To Do

• 1, RNN speech recognition (Tied-context-dependent-state and End-to-End)
• 2, Real environment noise cancellation(DNN-DAE/CNN-DAE/RNN-DAE: echo or reverberation)
• 3, Integrate the class information to HCLG fst for speech recognition
• 4, Multi-Mode features based VAD
• 5, DNN based Language identification and Speaker identification
• 6, Distant speech recognition(Reverberation, Mutli-microphones)
• 7, Voice conversation
• 8, Unbound activation function(Rectifier/Maxout/Pnorm) go-through searching method.
• 9, Sparse DNN
• 10, Neural network visulization
• 11, DAE+dropout

Technical Report To Write

• 1, DNN-DAE based noise cancellation --Mengyuan Zhao
• 2, Speech Rate DNN speech recognition --Shi Yin
• 3, CNN+fbank feature combination --Mian Wang
• 4, Uyghur low-resource acoustic model enhancement --Shi Yin
• 5, Uyghur 20h database release --

Papers To Read

• 1, Learned-Norm pooling for deep feedforward and recurrent neural networks

Task schedules

Summary

   --------------------------------------------------------------------------------------------------------
    Priority | Tasks name                    |      Status          |     Notions
   --------------------------------------------------------------------------------------------------------    
        1    | Bi-Softmax                    | ■■■□□□□□□□ | 1400h am training and problem fixing
   --------------------------------------------------------------------------------------------------------
        2    | RNN+DAE                       | □□□□□□□□□□ |
   --------------------------------------------------------------------------------------------------------

Speech Recognition

Multi-lingual Am training

Bi-Softmax

• Using two distinct softmax for English and Chinese data.
• Testing on 100h-Ch+100h-En, better performance observed.
• Now testing the source code on 1400h_8k data, but stange decoding results got.Need to further investigate.