“ASR Status Report 2016-12-05”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（5位用户的11个中间修订版本未显示）
第7行： 第7行：
 
|Yanqing Wang
 
|Yanqing Wang
 
||  
 
||  
−
*   
+
*  finish building GUI of one-class-SVM in Visual Studio
−
*
+
**[[媒体文件:One-class SVM.png|screenShot]]
−
*
+
* learn about deep learning
−
*  
+
 
||  
 
||  
−
*   
+
*  process data with the GUI of one-class-SVM
−
* 
+
 
|-
 
|-
  
第22行： 第20行：
 
|Hang Luo
 
|Hang Luo
 
||  
 
||  
−
*
+
* Experiments about fixed language model, find language information almost has no improve in ASR so far
−
*
+
* share one paper
−
*  
+
* Experiments on smaller language model,128 cells still has great result
−
* 
+
 
||  
 
||  
−
*  
+
* Try different language model or other direction
 +
* Paper about joint training
 
|-
 
|-
  
第49行： 第47行：
 
|Yixiang Chen   
 
|Yixiang Chen   
 
||  
 
||  
−
*   
+
*  According to the weighted filter replay test
−
*   
+
*  fitting warping function and use piecewise function for warping
 
||  
 
||  
−
*  
+
* complete replay test
 +
*  learning tensorfloat
 
|-
 
|-
−
 
  
  
第60行： 第58行：
 
|Lantian Li   
 
|Lantian Li   
 
||  
 
||  
−
*  
+
* LRE on AP16-OL7. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=574]
−
*
+
** The deep model still can not beat the i-vector model.
−
*
+
* Emotion challenge in CCPR. --> Done.
−
*  
+
* Replay detection. --> Done.
−
*
+
* Phd forum. --> Done
 
||  
 
||  
−
*
+
* LRE task.
−
*
+
* Write replay detection technical report.
 +
* oral PPT for APSIPA 2016.
 
|-
 
|-
−
 
  
  
第75行： 第73行：
 
|Zhiyuan Tang  
 
|Zhiyuan Tang  
 
||  
 
||  
−
*
+
* materials for PhD Dissertations,
−
*
+
* materials for paper accepted by TASLP;
 +
* language mask[http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=536]
 
||  
 
||  
−
*
+
* scans the TASLP paper for several times,
−
*  
+
* oral PPT for APSIPA 2016.
 
|-
 
|-
  

2016年12月7日 (三) 03:30的最后版本

Date People Last Week This Week
2016.11.28


Yanqing Wang
  • finish building GUI of one-class-SVM in Visual Studio
  • learn about deep learning
  • process data with the GUI of one-class-SVM
Hang Luo
  • Experiments about fixed language model, find language information almost has no improve in ASR so far
  • share one paper
  • Experiments on smaller language model,128 cells still has great result
  • Try different language model or other direction
  • Paper about joint training
Ying Shi
  • some work about kazak LM
  • DNN for Chao Xing
  • ml-book check to cvs
  • share two papers
  • some work about kazak lm(try to reduce the ppl)
  • paper about activation function
  • DNN for Chao Xing
Yixiang Chen
  • According to the weighted filter replay test
  • fitting warping function and use piecewise function for warping
  • complete replay test
  • learning tensorfloat
Lantian Li
  • LRE on AP16-OL7. [1]
    • The deep model still can not beat the i-vector model.
  • Emotion challenge in CCPR. --> Done.
  • Replay detection. --> Done.
  • Phd forum. --> Done
  • LRE task.
  • Write replay detection technical report.
  • oral PPT for APSIPA 2016.
Zhiyuan Tang
  • materials for PhD Dissertations,
  • materials for paper accepted by TASLP;
  • language mask[2]
  • scans the TASLP paper for several times,
  • oral PPT for APSIPA 2016.
Date People Last Week This Week
2016.11.28


Yanqing Wang
  • fighting with building GUI interface of one-class-SVM in Visual Studio but not finished yet
    • configure LibSVM toolkit in Visual Studio
    • implement the basic functions of LibSVM in Visual Studio
    • learn to create Qt project using Visual Studio and Qt Creator
  • finish building GUI interface of one-class-SVM in Visual Studio
  • make the GUI interface display dynamically with the change of given data
Hang Luo
  • Do experiments about joint learning including:
    • Fixed language model and give its information to speech model
    • Try smaller language model and find its result
  • Inter Speech paper talk about multi-task and highway connection
  • Continue to do experiments and read paper about language recognition model
Ying Shi
  • some work about kazak speech recognition
  • cnn visualization
  • paper reading
  • ML-book done
  • DNN with different activation function (Sigmoid Tanh Relu pnorm)
  • cnn visualization
Yixiang Chen
  • Continue replay detection (Freq-Weighting and Mel-Weighting).
  • Pooling replay data for UBM training
  • Continue replay detection (Change data set and Warping)
Lantian Li
  • LRE on AP16-OL7. [3]
    • 'StatisticsComponent'
    • The effect of Vad / Padding.
  • Replay detection.
    • performance-driven based Freq-Weighting
  • LRE task.
  • Freq-Warping and CNN-training.
Zhiyuan Tang
  • A speech named 'Deep Learning in Speech Recognition' in Chengdu;
  • Decoding with language mask seems helpless, not concluded.
  • use language mask in a proper way.
  • prepare materials for paper accepted by TASLP.
取自“http://index.cslt.org/mediawiki/index.php?title=ASR_Status_Report_2016-12-05&oldid=24627”