“ASR:2015-04-08”版本间的差异
来自cslt Wiki
（→Text Processing） |
|||
| 第41行： | 第41行： | ||
* similar word extension in FST | * similar word extension in FST | ||
:* check the formula using Bayes and experiment | :* check the formula using Bayes and experiment | ||
| + | :* add more test data | ||
| + | :* test the baseline(no weight) and different weight method | ||
====RNN LM==== | ====RNN LM==== | ||
| 第49行： | 第51行： | ||
====W2V based doc classification==== | ====W2V based doc classification==== | ||
| − | * | + | * reproducible test using English data |
| − | * | + | * Code new version spherical word vector. |
| + | * Accomplish movMF model | ||
===Translation=== | ===Translation=== | ||
| 第58行： | 第61行： | ||
===Sparse NN in NLP=== | ===Sparse NN in NLP=== | ||
* prepare the ACL | * prepare the ACL | ||
| − | :* | + | :* test result is ok now[http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=lr&step=view_request&cvssid=344]. |
| − | :* | + | :* find the new direction. |
| − | + | ||
===online learning=== | ===online learning=== | ||
* data is ready.prepare the ACL paper | * data is ready.prepare the ACL paper | ||
| − | :* | + | :* finish some test. |
| − | :* | + | :* test the result on different time. |
| + | |||
| + | ===relation classifier=== | ||
| + | * check code and find the problem that result is different on sigmoid and tanh | ||
2015年4月8日 (三) 10:50的最后版本
Speech Processing
AM development
Environment
- grid-11 often shut down automatically, too slow computation speed.
RNN AM
- details at http://liuc.cslt.org/pages/rnnam.html
- tuning parameters on monophone NN
- run using wsj,MPE
Mic-Array
- investigate alpha parameter in time domian and frquency domain
- ALPHA>=0
Convolutive network
- HOLD
- CNN + DNN feature fusion
RNN-DAE(Deep based Auto-Encode-RNN)
- HOLD -Zhiyong
- http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=261
Speaker ID
- DNN-based sid --Yiye
- Decode --Yiye
- http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=327
Ivector based ASR
- http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?step=view_request&cvssid=340
- Ivector dimention is smaller, performance is better
- Augument to hidden layer is better than input layer
- train on wsj(testbase dev93+evl92)
Text Processing
tag LM
- similar word extension in FST
- check the formula using Bayes and experiment
- add more test data
- test the baseline(no weight) and different weight method
RNN LM
- rnn
- code the character-lm using Theano
- lstm+rnn
- check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)
W2V based doc classification
- reproducible test using English data
- Code new version spherical word vector.
- Accomplish movMF model
Translation
- v5.0 demo released
- cut the dict and use new segment-tool
Sparse NN in NLP
- prepare the ACL
- test result is ok now[1].
- find the new direction.
online learning
- data is ready.prepare the ACL paper
- finish some test.
- test the result on different time.
relation classifier
- check code and find the problem that result is different on sigmoid and tanh
