ASR:2015-06-15

Speech Processing

AM development

RNN AM

• morpheme RNN-zhiyuan
• RNN MPE --zhiyuan and xuewei

Mic-Array

• hold
• Change the prediction from fbank to spectrum features
• investigate alpha parameter in time domian and frquency domain
• ALPHA>=0, using data generated by reverber toolkit
• consider theta
• compute EER with kaldi

RNN-DAE(Deep based Auto-Encode-RNN)

• hold
• deliver to mengyuan

• http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=261

Speaker ID

• DNN-based sid --Lantian

• http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=327

Ivector&Dvector based ASR

• hold --Tian Lan
• Cluster the speakers to speaker-classes, then using the distance or the posterior-probability as the metric
• dark-konowlege using i-vector
• train on wsj(testbase dev93+evl92)

• --hold

Dark knowledge

• test random last output layer when train MPE--zhiyuan

bilingual recognition

• imbalance dataset(10h,100h and 1400h) to train without share--Zhiyong and mengyuan
• record utterances mixed with Chinese and English

language vector

• hold --xuewei

DNN MPE

• random the last layer, train MPE--zhiyuan

Text Processing

RNN LM

• character-lm rnn(hold)
• lstm+rnn

• check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)

W2V based document classification

• APSIPA paper
• CNN adapt to resolve the low resource problem

Pair-wise LM

• draft paper of journal

Order representation

• modify the objective function(hold)
• sup-sampling method to solve the low frequence word(hold)
• journal paper

binary vector

• nips paper

Stochastic ListNet

• done

relation classifier

• done

plan to do

• combine LDA with neural network