“2022-01-17”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
（以“{| class="wikitable" !People !! This Week !! Next Week !! Task Tracking (<font color="red">DeadLine</font>) |- |- |Dong Wang || * || * || * |- |- |Yunqi Cai ||...”为内容创建页面）
 
 
（12位用户的16个中间修订版本未显示）
第5行： 第5行：
 
|Dong Wang
 
|Dong Wang
 
||  
 
||  
−
*
+
* OLR workshop preparation
 +
* CNSRC plan: metric design
 +
* Capsulate review: emotion recognition
 +
* Visual-audio SID experiment plan design
 
||
 
||
−
*
+
* Keep on multi-modal review & information fusion
 
||
 
||
 
*   
 
*   
第16行： 第19行：
 
|Yunqi Cai
 
|Yunqi Cai
 
||   
 
||   
−
*
+
*Low-light image simulation and data preparation
 +
*capsule network review
 +
*Research plan guidance for interns
 +
*Conceptual model of intelligent sensor
 
||
 
||
 
*  
 
*  
第39行： 第45行：
 
|Ying Shi
 
|Ying Shi
 
||  
 
||  
−
* Speech engrave (text attend speech)
+
* Analysis FA of Speech engrave
−
* attention visualization [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=shiying&step=view_request&cvssid=829 here]
+
* Speech engrave with position embedding [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=shiying&step=view_request&cvssid=829 here]
 
||
 
||
−
* Speech engrave v2
+
* Analysis Speech engrave with position embedding
 
||
 
||
 
*   
 
*   
第51行： 第57行：
 
|Haoran Sun
 
|Haoran Sun
 
||  
 
||  
−
*  
+
* experiments for rebuttal
 +
* update CycleFlow code
 +
* AutoVC with cycle loss
 
||
 
||
−
*  
+
* More training and evaluation on AutoVC
 +
* More attempts on cycle loss
 
||
 
||
 
*   
 
*   
第62行： 第71行：
 
|Chen Chen
 
|Chen Chen
 
||  
 
||  
−
* Finish course thesis
+
* Figure out the influence of symbol & feature
−
* Test UASR with different phone symbol
+
* Test w2v2.0 feats with LSTM in a supervised way
 
||
 
||
−
* Figure out reason of the influence of symbol
+
* Test the influence of feature & preprocessing & training epochs
−
* Test w2v2.0 feats with other NN instead of GAN
+
 
||
 
||
 
*   
 
*   
第75行： 第83行：
 
|Pengqi Li
 
|Pengqi Li
 
||   
 
||   
−
*  
+
* Read Paper (RELAX,RISE); Finish Demo about RISE
 +
* Evaluate the importance maps  with Deletion and Insertion
 
||
 
||
−
*  
+
* Test MFCC
 +
* Finish RELAX
 
||
 
||
 
*   
 
*   
第86行： 第96行：
 
|Weida Liang
 
|Weida Liang
 
||   
 
||   
−
* Finish wav2vec+decoder model and do self-evaluation test
+
* Paper sharing on any-to-any VC
−
* evaluate the scale of network
+
* Finish training wav2vec model
 +
* Ongoing wav2vec+AE model
 
||
 
||
−
* Finish the never-before seen test
+
* Finish all model training
−
* Paper sharing on Wednesday
+
* Test and comparison analysis
 
||
 
||
 
*   
 
*   
第98行： 第109行：
 
|Zixi Yan
 
|Zixi Yan
 
||   
 
||   
−
* Thchs30 speech recognition using Kaldi
+
* Write code for extracting wav2vec features in Kaldi
−
* Try using the WAV2VEc feature in Kaldi speech recognition, modify make_mfcc.sh
+
 
||
 
||
 
*
 
*
第109行： 第119行：
 
|Sirui Li
 
|Sirui Li
 
||   
 
||   
−
* Compare the effect of wav2vec parameter tuning
+
* Extract MFCC features for GAN
−
* Prepare group report
+
* Write a project application
 
||
 
||
−
* Extract Chinese features for GAN
+
* Go on GAN experiment
 
||
 
||
 
*   
 
*   
第121行： 第131行：
 
|Haoyu Jiang
 
|Haoyu Jiang
 
||  
 
||  
−
* Calculate the acc of each segment score
+
*Complete the data partition
 +
*Prepare group report
 
||
 
||
−
* Complete data merge
+
*
 
||
 
||
 
*   
 
*   
第132行： 第143行：
 
|Ruihai Hou
 
|Ruihai Hou
 
||  
 
||  
−
*  
+
* Read the UIS-RNN paper
 +
* Run some tests on speaker diarization
 
||
 
||
−
*  
+
* Try to train the UIS-RNN model
 
||
 
||
 
*   
 
*   
第142行： 第154行：
 
|Renmiao Chen
 
|Renmiao Chen
 
||  
 
||  
−
* Finish course task
+
* Do three experiments for PMI.
 +
* Do something for programming TA.
 
||
 
||
−
* Finish pmi
+
* Finish experiments for PMI.
 
|-
 
|-
  
 
|}
 
|}

2022年1月17日 (一) 10:59的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • OLR workshop preparation
  • CNSRC plan: metric design
  • Capsulate review: emotion recognition
  • Visual-audio SID experiment plan design
  • Keep on multi-modal review & information fusion
Yunqi Cai
  • Low-light image simulation and data preparation
  • capsule network review
  • Research plan guidance for interns
  • Conceptual model of intelligent sensor
Lantian Li
  • Polish CNCSRC plan
  • Write hard trials paper
  • Finish hard trials paper
Ying Shi
  • Analysis FA of Speech engrave
  • Speech engrave with position embedding here
  • Analysis Speech engrave with position embedding
Haoran Sun
  • experiments for rebuttal
  • update CycleFlow code
  • AutoVC with cycle loss
  • More training and evaluation on AutoVC
  • More attempts on cycle loss
Chen Chen
  • Figure out the influence of symbol & feature
  • Test w2v2.0 feats with LSTM in a supervised way
  • Test the influence of feature & preprocessing & training epochs
Pengqi Li
  • Read Paper (RELAX,RISE); Finish Demo about RISE
  • Evaluate the importance maps with Deletion and Insertion
  • Test MFCC
  • Finish RELAX
Weida Liang
  • Paper sharing on any-to-any VC
  • Finish training wav2vec model
  • Ongoing wav2vec+AE model
  • Finish all model training
  • Test and comparison analysis
Zixi Yan
  • Write code for extracting wav2vec features in Kaldi
Sirui Li
  • Extract MFCC features for GAN
  • Write a project application
  • Go on GAN experiment
Haoyu Jiang
  • Complete the data partition
  • Prepare group report
Ruihai Hou
  • Read the UIS-RNN paper
  • Run some tests on speaker diarization
  • Try to train the UIS-RNN model
Renmiao Chen
  • Do three experiments for PMI.
  • Do something for programming TA.
  • Finish experiments for PMI.
取自“http://index.cslt.org/mediawiki/index.php?title=2022-01-17&oldid=37660”