“2021-05-10”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(5位用户的7个中间修订版本未显示)
第6行: 第6行:
 
|Dong Wang
 
|Dong Wang
 
||  
 
||  
*
+
* AI-100 completed
 +
* Review the LSTF model
 
||
 
||
*
+
* More investigation on the LSTF model
 
||
 
||
 
*   
 
*   
第17行: 第18行:
 
|Yunqi Cai
 
|Yunqi Cai
 
||   
 
||   
*
+
*review papers about Infrared face recognition
 +
*investigation on the MG-xvector model
 
||
 
||
 
*  
 
*  
第28行: 第30行:
 
|Lantian Li
 
|Lantian Li
 
||  
 
||  
*  
+
* Human test.
*  
+
* Three standards.
 
||
 
||
*  
+
* CN-Celeb data/response.
 
||
 
||
 
*   
 
*   
第51行: 第53行:
 
|Haoran Sun
 
|Haoran Sun
 
||  
 
||  
*  
+
* gmvae for single speaker
*  
+
* it seems hard to implement this on multi-spkeakers
 
||
 
||
*  
+
* find similar models without too many parameters
 
*  
 
*  
 
||
 
||
第65行: 第67行:
 
|Jiao Han
 
|Jiao Han
 
||  
 
||  
*  
+
* Verification about vectors'uniform distribution and MG taining.
 
||
 
||
 
*  
 
*  
第76行: 第78行:
 
|Yang Zhang
 
|Yang Zhang
 
||  
 
||  
*  
+
* cnc check
 
||
 
||
 
*  
 
*  

2021年5月10日 (一) 06:21的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • AI-100 completed
  • Review the LSTF model
  • More investigation on the LSTF model
Yunqi Cai
  • review papers about Infrared face recognition
  • investigation on the MG-xvector model
Lantian Li
  • Human test.
  • Three standards.
  • CN-Celeb data/response.
Ying Shi
  • test agc and denoise of webrtc
Haoran Sun
  • gmvae for single speaker
  • it seems hard to implement this on multi-spkeakers
  • find similar models without too many parameters
Jiao Han
  • Verification about vectors'uniform distribution and MG taining.
Yang Zhang
  • cnc check
Di Wang
  • Begin the first round of evaluation activities.
  • Screening out the new "trials".
  • Investigate the crowd test platform.
  • Carry out evaluation activities.
Tiankai Zhi
Chen Chen
  • modify spon web front-end & back-end
  • read paper about MetricGAN & learn speechbrain's recipe
  • run speechbrain's recipe successfully