“ASR:2015-06-15”版本间的差异
来自cslt Wiki
（→RNN AM） |
（→Speech Processing） |
||
| 第6行： | 第6行： | ||
==== RNN AM==== | ==== RNN AM==== | ||
*morpheme RNN-zhiyuan | *morpheme RNN-zhiyuan | ||
| + | *RNN MPE --zhiyuan and xuewei | ||
==== Mic-Array ==== | ==== Mic-Array ==== | ||
| 第21行： | 第22行： | ||
===Speaker ID=== | ===Speaker ID=== | ||
| − | * DNN-based sid -- | + | * DNN-based sid --Lantian |
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=327 | :* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=327 | ||
===Ivector&Dvector based ASR=== | ===Ivector&Dvector based ASR=== | ||
| − | * | + | * hold --Tian Lan |
* Cluster the speakers to speaker-classes, then using the distance or the posterior-probability as the metric | * 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) | * train on wsj(testbase dev93+evl92) | ||
| + | :*--hold | ||
===Dark knowledge=== | ===Dark knowledge=== | ||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
* test random last output layer when train MPE--zhiyuan | * test random last output layer when train MPE--zhiyuan | ||
===bilingual recognition=== | ===bilingual recognition=== | ||
| − | * | + | * imbalance dataset(10h,100h and 1400h) to train without share--Zhiyong and mengyuan |
| − | + | * record utterances mixed with Chinese and English | |
===language vector=== | ===language vector=== | ||
| − | * | + | * hold --xuewei |
| + | |||
| + | ===DNN MPE=== | ||
| + | * random the last layer, train MPE--zhiyuan | ||
==Text Processing== | ==Text Processing== | ||
2015年6月17日 (三) 06:33的版本
Speech Processing
AM development
Environment
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
Speaker ID
- DNN-based sid --Lantian
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
