“ASR:2015-03-09”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
（→‎Sparse NN in NLP）
Lr（讨论 | 贡献）
（→‎Text Processing）
 
（2位用户的4个中间修订版本未显示）
第3行： 第3行：
  
 
==== Environment ====
 
==== Environment ====
−
* grid-11 often shutdown automatically, too slow computation speed.
+
* grid-11 often shut down automatically, too slow computation speed.
−
* buy a new 800W power -- Xuewei
+
* GPU has being repired.--Xuewei
  
 
==== RNN AM====
 
==== RNN AM====
 
* details at http://liuc.cslt.org/pages/rnnam.html
 
* details at http://liuc.cslt.org/pages/rnnam.html
−
* triphone one state based RNN?
+
* triphone one state based RNN ?--Liu Chao
  
 
==== Mic-Array ====
 
==== Mic-Array ====
−
* the technical report is done.
 
 
* reproduce environment for interspeech
 
* reproduce environment for interspeech
 +
* alpha parameter in Lasso
  
 
====Dropout & Maxout & rectifier ====
 
====Dropout & Maxout & rectifier ====
 
* HOLD
 
* HOLD
 
* Need to solve the too small learning-rate problem
 
* Need to solve the too small learning-rate problem
−
* 20h small scale sparse dnn with rectifier. --Chao liu
+
* 20h small scale sparse dnn with rectifier. --Mengyuan
 
* 20h small scale sparse dnn with Maxout/rectifier based on weight-magnitude-pruning. --Mengyuan Zhao
 
* 20h small scale sparse dnn with Maxout/rectifier based on weight-magnitude-pruning. --Mengyuan Zhao
  
 
====Convolutive network====
 
====Convolutive network====
 
* Convolutive network(DAE)
 
* Convolutive network(DAE)
−
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=311
+
* HOLD
 
:* Technical report writing, Mian Wang, Yiye Lin, Shi Yin, Mengyuan Zhao
 
:* Technical report writing, Mian Wang, Yiye Lin, Shi Yin, Mengyuan Zhao
 
:* reproduce experiments -- Yiye
 
:* reproduce experiments -- Yiye
第32行： 第32行：
  
 
====RNN-DAE(Deep based Auto-Encode-RNN)====
 
====RNN-DAE(Deep based Auto-Encode-RNN)====
−
* HOLD
+
* HOLD -Zhiyong
 
* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=261
 
* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=261
−
 
−
====VAD====
 
−
* DAE
 
−
* Technical report done. -- Shi Yin
 
  
 
====Speech rate training====
 
====Speech rate training====
 
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=268
 
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=268
−
:* Technical report to draft. Xiangyu Zeng, Shi Yin
+
:* Technical report to draft.-- Xiangyu Zeng, Shi Yin
 
:* Prepare for NCMSSC
 
:* Prepare for NCMSSC
−
 
−
====Confidence====
 
−
* HOLD
 
−
* Reproduce the experiments on fisher dataset.
 
−
* Use the fisher DNN model to decode all-wsj dataset
 
−
* preparing scoring for puqiang data
 
  
 
====Neural network visulization====
 
====Neural network visulization====
 
* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=324
 
* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=324
−
* Technical report writing, Mian Wang.
+
* Technical report writing --Mian Wang.
  
 
===Speaker ID===   
 
===Speaker ID===   
−
:* DNN-based sid
+
:* DNN-based sid --Yiye
 
:* 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 based ASR===
 +
:* Ivector dimention is smaller, performance is better
 +
:* Augument to hidden layer is better than input layer
  
 
==Text Processing==
 
==Text Processing==
第65行： 第57行：
 
====Domain specific LM====
 
====Domain specific LM====
 
* LM2.X
 
* LM2.X
−
:* mix the sougou2T-lm,kn-discount(done)
 
 
:* train a large lm using 25w-dict.(hanzhenglong/wxx)
 
:* train a large lm using 25w-dict.(hanzhenglong/wxx)
 
::* v2.0c filter the useless word.(next week)
 
::* v2.0c filter the useless word.(next week)
 
::* set the test set for new word (hold)
 
::* set the test set for new word (hold)
 +
:* prepare the wiki data: entity list.
  
 
====tag LM====
 
====tag LM====
第78行： 第70行：
 
====RNN LM====
 
====RNN LM====
 
*rnn
 
*rnn
−
:* discuss the rnn-lstm lm
+
:* the input and output is word embedding and add some token information like NER..
 +
:* map the word to character and train the lm.
 
*lstm+rnn
 
*lstm+rnn
 
:* check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)
 
:* check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)
第87行： 第80行：
 
* data prepare.(hold)
 
* data prepare.(hold)
 
====Knowledge vector====
 
====Knowledge vector====
−
* paper is done, submitted ACL
+
* make a report on Monday
 +
 
 
===Translation===
 
===Translation===
  
第95行： 第89行：
 
===Sparse NN in NLP===
 
===Sparse NN in NLP===
 
* prepare the ACL
 
* prepare the ACL
−
 
+
:* check the code to find the problem .
 +
:* increase the dimension
 +
:* use different test set.
 
===QA===
 
===QA===
 
====improve fuzzy match====
 
====improve fuzzy match====
第101行： 第97行：
 
===online learning===
 
===online learning===
 
* data is ready.prepare the ACL paper
 
* data is ready.prepare the ACL paper
−
====context framework====
+
:* prepare sougouQ data and test it using current online learning method
−
* code for demo
+
====framework====
−
:* demo is done
+
* extract the module
−
 
+
:* extract the context module ,search module,entity recognize module and common module.
 +
:* define the inference in different modules
 +
* composite module
  
 +
====leftover problem====
 
* new inter will install SEMPRE
 
* new inter will install SEMPRE

2015年3月16日 (一) 01:11的最后版本

Speech Processing

AM development

Environment

  • grid-11 often shut down automatically, too slow computation speed.
  • GPU has being repired.--Xuewei

RNN AM

Mic-Array

  • reproduce environment for interspeech
  • alpha parameter in Lasso

Dropout & Maxout & rectifier

  • HOLD
  • Need to solve the too small learning-rate problem
  • 20h small scale sparse dnn with rectifier. --Mengyuan
  • 20h small scale sparse dnn with Maxout/rectifier based on weight-magnitude-pruning. --Mengyuan Zhao

Convolutive network

  • Convolutive network(DAE)
  • HOLD
  • Technical report writing, Mian Wang, Yiye Lin, Shi Yin, Mengyuan Zhao
  • reproduce experiments -- Yiye

DNN-DAE(Deep Auto-Encode-DNN)

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

Speech rate training

Neural network visulization

Speaker ID

Ivector based ASR

  • Ivector dimention is smaller, performance is better
  • Augument to hidden layer is better than input layer

Text Processing

LM development

Domain specific LM

  • LM2.X
  • train a large lm using 25w-dict.(hanzhenglong/wxx)
  • v2.0c filter the useless word.(next week)
  • set the test set for new word (hold)
  • prepare the wiki data: entity list.

tag LM

  • Tag Lm(JT)
  • error check
  • similar word extension in FST
  • add the experiment to tag-lm paper.

RNN LM

  • rnn
  • the input and output is word embedding and add some token information like NER..
  • map the word to character and train the lm.
  • lstm+rnn
  • check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)

Word2Vector

W2V based doc classification

  • data prepare.(hold)

Knowledge vector

  • make a report on Monday

Translation

  • v5.0 demo released
  • cut the dict and use new segment-tool

Sparse NN in NLP

  • prepare the ACL
  • check the code to find the problem .
  • increase the dimension
  • use different test set.

QA

improve fuzzy match

  • add Synonyms similarity using MERT-4 method(hold)

online learning

  • data is ready.prepare the ACL paper
  • prepare sougouQ data and test it using current online learning method

framework

  • extract the module
  • extract the context module ,search module,entity recognize module and common module.
  • define the inference in different modules
  • composite module

leftover problem

  • new inter will install SEMPRE
取自“http://cslt.org/mediawiki/index.php?title=ASR:2015-03-09&oldid=14233”