“ASR:2015-01-12”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
（→‎tag LM）
（→‎RNN-DAE(RNN based Atuo-Encode)）
 
（2位用户的4个中间修订版本未显示）
第4行： 第4行：
 
==== Environment ====
 
==== Environment ====
 
* May gpu760 of grid-14 be something wrong. To be exchanged.
 
* May gpu760 of grid-14 be something wrong. To be exchanged.
 +
* grid-2 can not be booted
 +
* grid-10 should replac the CPU fan.
  
 
==== Sparse DNN ====
 
==== Sparse DNN ====
 
* details at http://liuc.cslt.org/pages/sparse.html
 
* details at http://liuc.cslt.org/pages/sparse.html
−
* need to test clean data
 
−
* MPE training to be continue
 
  
 
==== RNN AM====
 
==== RNN AM====
第16行： 第16行：
 
====Dropout & Maxout & retifier ====
 
====Dropout & Maxout & retifier ====
 
* Drop out
 
* Drop out
−
:* Change the test data to more noisy data, to verify the effectiveness of dropout.
 
  
−
* MaxOut && P-norm
+
* MaxOut && P-norm(+)
 
:* Need to solve the too small learning-rate problem
 
:* Need to solve the too small learning-rate problem
 
:** Add one normalization layer after the pnorm-layer
 
:** Add one normalization layer after the pnorm-layer
第27行： 第26行：
 
* 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
 
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=311
 +
 
:* Feature extractor  
 
:* Feature extractor  
 
:** Combined with raw features, better performance obsearved.
 
:** Combined with raw features, better performance obsearved.
−
:** Technical report to draft, Yiye Lin, Shi Yin, Menyuan Zhao and Mian Wang
+
:** Technical report to draft, Yiye Lin, Shi Yin, Menyuan Zhao and Mian Wang(+)
−
:* To test real enviroment echo.
+
:* To test real enviroment echo.(+)
  
 
====DNN-DAE(Deep Atuo-Encode-DNN)====
 
====DNN-DAE(Deep Atuo-Encode-DNN)====
−
:* test on XinWenLianBo music. results on
+
:* Technical report to draft, Mengyuan Zhao and Zhiyong Zhang.
−
:** http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhaomy&step=view_request&cvssid=318
+
−
:** Technical report to draft, Mengyuan Zhao and Zhiyong Zhang.
+
 
:* To test real enviroment echo.
 
:* To test real enviroment echo.
 +
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=318
 +
:* XueWei will reproduce the experiments.
 +
 +
====RNN-DAE(Deep based Atuo-Encode- RNN)====
 +
:* http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=zhangzy&step=view_request&cvssid=261
  
 
====VAD====
 
====VAD====
 
* Harmonics and Teager energy features done.
 
* Harmonics and Teager energy features done.
−
* Model to be trained.
+
* Frame energy Model training done.
 +
* Other features model to be train.
  
 
====Speech rate training====
 
====Speech rate training====
第66行： 第70行：
  
 
===Voice Conversion===
 
===Voice Conversion===
−
* Yiye is reading materials(+)
+
* Yiye is reading materials
 +
* HOLD
  
  
第106行： 第111行：
 
:*  
 
:*  
 
:* Modify the object function and training process.
 
:* Modify the object function and training process.
 +
 
====relation====
 
====relation====
 
====Character to word====
 
====Character to word====

2015年1月16日 (五) 10:31的最后版本

Speech Processing

AM development

Environment

  • May gpu760 of grid-14 be something wrong. To be exchanged.
  • grid-2 can not be booted
  • grid-10 should replac the CPU fan.

Sparse DNN

RNN AM

Dropout & Maxout & retifier

  • Drop out
  • MaxOut && P-norm(+)
  • Need to solve the too small learning-rate problem
    • Add one normalization layer after the pnorm-layer
    • Add L2-norm upper bound
  • hold

Convolutive network

  • Convolutive network(DAE)
  • Feature extractor
    • Combined with raw features, better performance obsearved.
    • Technical report to draft, Yiye Lin, Shi Yin, Menyuan Zhao and Mian Wang(+)
  • To test real enviroment echo.(+)

DNN-DAE(Deep Atuo-Encode-DNN)

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

VAD

  • Harmonics and Teager energy features done.
  • Frame energy Model training done.
  • Other features model to be train.

Speech rate training

Confidence

  • Reproduce the experiments on fisher dataset.
  • Use the fisher DNN model to decode all-wsj dataset
  • preparing scoring for puqiang data
  • HOLD

Neural network visulization

Speaker ID

Language ID

Voice Conversion

  • Yiye is reading materials
  • HOLD


Text Processing

LM development

Domain specific LM

  • LM2.1
  • mix the sougou2T-lm,kn-discount continue
  • train a large lm using 25w-dict.(hanzhenglong/wxx)
  • data pre-processing("this week")
  • new dict.
  • dongxu help zhenglong with large dictionary.

tag LM

  • Tag Lm
  • tag Probability should test add the weight(hanzhenglong) and handover to hanzhenglong ("this month")
  • paper
  • paper submit this week.
  • similar word extension in FST
  • find similarity word using word2vec,word vector is training.
  • set the weight for word

RNN LM

  • rnn
  • test wer RNNLM on Chinese data from jietong-data
  • generate the ngram model from rnnlm and test the ppl with different size txt.
  • lstm+rnn
  • check the lstm-rnnlm code about how to Initialize and update learning rate.(hold)

Word2Vector

W2V based doc classification

  • data prepare.

Knowledge vector

  • Knowledge vector
  • Make a proper test set.
  • Modify the object function and training process.

relation

Character to word

  • Character to word conversion(hold)

Translation

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

QA

improve fuzzy match

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

improve lucene search

  • add more feature to improve search.
  • POS, NER ,tf ,idf ..

context framework

  • code for organization

query normalization

  • using NER to normalize the word
  • new inter will install SEMPRE
取自“http://cslt.org/mediawiki/index.php?title=ASR:2015-01-12&oldid=13560”