“2022-01-10”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
第88行： 第88行：
 
|Weida Liang
 
|Weida Liang
 
||   
 
||   
−
*  
+
* Finish wav2vec+decoder model and do self-evaluation test
 +
* evaluate the scale of network
 
||
 
||
−
*  
+
* Finish the never-before seen test
 
||
 
||
 
*   
 
*   

2022年1月10日 (一) 11:04的版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Finish CNSRC challenge plan
  • Deal with ICASSP response
  • Keep on survey for multi-modal information fusion
  • Keep on multi-modal technique review
Yunqi Cai
  • Capsnet generative model investigation
Lantian Li
  • Push CNCSRC plan
  • Make Real AM rebuttal
  • Go on hard trials paper
Ying Shi
  • Speech engrave (text attend speech)
  • attention visualization here
  • Speech engrave v2
Haoran Sun
Chen Chen
  • Finish course thesis
  • Test UASR with different phone symbol
  • Figure out reason of the influence of symbol
  • Test w2v2.0 feats with other NN instead of GAN
Pengqi Li
Weida Liang
  • Finish wav2vec+decoder model and do self-evaluation test
  • evaluate the scale of network
  • Finish the never-before seen test
Zixi Yan
  • Thchs30 speech recognition using Kaldi
  • Try using the WAV2VEc feature in Kaldi speech recognition, modify make_mfcc.sh
Sirui Li
  • Compare the effect of wav2vec parameter tuning
  • Prepare group report
  • Extract Chinese features for GAN
Haoyu Jiang
  • Calculate the acc of each segment score
  • Complete data merge
Ruihai Hou
Renmiao Chen
取自“http://index.cslt.org/mediawiki/index.php?title=2022-01-10&oldid=37581”