“ASR Status Report 2016-10-24”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（1位用户的4个中间修订版本未显示）
第6行： 第6行：
  
 
|-
 
|-
−
| rowspan="4"|2016.10.24
+
| rowspan="5"|2016.10.24
 
|Hang Luo   
 
|Hang Luo   
 
||  
 
||  
−
*   
+
*  Read paper about joint training
 +
*  Read paper of Inter Speech
 
||  
 
||  
−
*   
+
*  Know the source code of joint training
 +
*  Joint training on speech and language
 
|-
 
|-
  
第21行： 第23行：
 
*  visualization job of DNN model report for weekly reading(one paper)
 
*  visualization job of DNN model report for weekly reading(one paper)
 
||  
 
||  
−
*  run CNN model ;sample frame training ; paper and if the resualt of sample frame training is amazing visualized it;ML-book  
+
*  run CNN model ;sample frame training ; paper and if the result of sample frame training is amazing visualized it;ML-book  
 +
|-
 +
 
 +
 
 +
|-
 +
|Yixiang Chen 
 +
||
 +
* channel detection.
 +
* ML-book
 +
||
 +
*  Joint-training
 
|-
 
|-
  
第29行： 第41行：
 
|Lantian Li   
 
|Lantian Li   
 
||  
 
||  
−
*
+
* Replay spoofing -- channel detection. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=571]
 +
* Joint-training on TASLP. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=573]
 
||  
 
||  
−
*
+
* Joint-training on TASLP.
 +
* Joint-training on spk and lan.
 
|-
 
|-
  

2016年11月1日 (二) 03:20的最后版本

Date People Last Week This Week


2016.10.24 Hang Luo
  • Read paper about joint training
  • Read paper of Inter Speech
  • Know the source code of joint training
  • Joint training on speech and language
Ying Shi
  • visualization job of DNN model report for weekly reading(one paper)
  • run CNN model ;sample frame training ; paper and if the result of sample frame training is amazing visualized it;ML-book
Yixiang Chen
  • channel detection.
  • ML-book
  • Joint-training
Lantian Li
  • Replay spoofing -- channel detection. [1]
  • Joint-training on TASLP. [2]
  • Joint-training on TASLP.
  • Joint-training on spk and lan.
Zhiyuan Tang
  • report for Weekly Reading, a brief lecture of NN in ASR for Tonghuashun members;
  • additinal experiments of joint learning (speech & spk) for taslp (multi-task, ivector as part of input)[3];
  • gru as the basic component for joint learning (speech & language)[4].
  • O-COCOSDA 16, Bali.
取自“http://cslt.org/mediawiki/index.php?title=ASR_Status_Report_2016-10-24&oldid=23291”