“2023-10-09”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(8位用户的10个中间修订版本未显示)
第6行: 第6行:
 
|Dong Wang
 
|Dong Wang
 
||  
 
||  
*
+
* Speech Information Processing, chapt 3 & 4
 +
* Three public presentations
 +
* AI book amendment
 +
 
 
||
 
||
 
*
 
*
第17行: 第20行:
 
|Lantian Li
 
|Lantian Li
 
||
 
||
* GPU status[https://z1et6d3xtb.feishu.cn/docx/Mxyzdnz4EoR8lVxX800cKcWinUg?pre_pathname=%2Fdrive%2Fhome%2F]
+
* GPU status [https://z1et6d3xtb.feishu.cn/wiki/XGcGwRK5viJmpRkjH9AczIhynCh]
 
* APT-group (Beam Search)
 
* APT-group (Beam Search)
 
* AI course
 
* AI course
第31行: 第34行:
 
|Ying Shi
 
|Ying Shi
 
||  
 
||  
*
+
* Speech separation with a simple keyword(without asr loss)
 +
* Speaker attributed speech separation (in progress)
 +
* Retrain waveGAN with ASR feature(in progress)
 
||
 
||
 
*  
 
*  
第42行: 第47行:
 
|Zhenghai You
 
|Zhenghai You
 
||  
 
||  
*  
+
* Some experiments in Voxceleb and Librispeech (in progress)
 +
* paper reading about ASR
 
||
 
||
 
*  
 
*  
第53行: 第59行:
 
|Chen Chen
 
|Chen Chen
 
||  
 
||  
*  
+
* Experiments of Cycleflow-unit-VTS [https://z1et6d3xtb.feishu.cn/docx/QV7pdhunMoSeAqxR0JzcSOuwnWf?from=from_copylink]
 +
||
 +
* More experiments
 +
* NCMMSC paper: CNVSRC baseline system
 
||
 
||
 
*  
 
*  
||
 
 
 
|-
 
|-
  
第75行: 第82行:
 
|Zehua Liu
 
|Zehua Liu
 
||  
 
||  
*  
+
* VTS paper reading and baseline
 +
* SadTalker begin
 
||
 
||
 
*  
 
*  
第86行: 第94行:
 
|Pengqi Li
 
|Pengqi Li
 
||   
 
||   
*  
+
* Phone's discrimination in deep SID [https://z1et6d3xtb.feishu.cn/docx/TkIZdRT9NoZ2dSxZth5ci3hxnYe]
 
||
 
||
 
*
 
*
第120行: 第128行:
 
|Zhenyu Zhou
 
|Zhenyu Zhou
 
||  
 
||  
*
+
*Aug+dann[https://z1et6d3xtb.feishu.cn/docx/Jbgnd8W5Uonh63xyaL8c2tYZnuf]
 
||
 
||
 
*  
 
*  
第143行: 第151行:
 
||  
 
||  
 
* beam search paper reading
 
* beam search paper reading
* pure beam search using variance as weight[https://z1et6d3xtb.feishu.cn/docx/Mxyzdnz4EoR8lVxX800cKcWinUg]
+
* pure beam search using variance as weight [https://z1et6d3xtb.feishu.cn/docx/Mxyzdnz4EoR8lVxX800cKcWinUg]
 
||
 
||
 
*  
 
*  
第156行: 第164行:
 
* Crowd sound event detection test
 
* Crowd sound event detection test
 
* EER test
 
* EER test
* Model size analyze[https://z1et6d3xtb.feishu.cn/docx/X8bpdvuMHoJ76LxhAPNcxfIPnee]
+
* Model size analyze [https://z1et6d3xtb.feishu.cn/docx/X8bpdvuMHoJ76LxhAPNcxfIPnee]
 
||
 
||
 
*
 
*
第178行: 第186行:
 
|Yang Wei
 
|Yang Wei
 
||  
 
||  
*
+
* Prepare FunASR for shengyuanzhineng
 
||
 
||
 
*
 
*

2023年10月9日 (一) 10:55的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Speech Information Processing, chapt 3 & 4
  • Three public presentations
  • AI book amendment
Lantian Li
  • GPU status [1]
  • APT-group (Beam Search)
  • AI course
  • CIPS defense
Ying Shi
  • Speech separation with a simple keyword(without asr loss)
  • Speaker attributed speech separation (in progress)
  • Retrain waveGAN with ASR feature(in progress)
Zhenghai You
  • Some experiments in Voxceleb and Librispeech (in progress)
  • paper reading about ASR
Chen Chen
  • Experiments of Cycleflow-unit-VTS [2]
  • More experiments
  • NCMMSC paper: CNVSRC baseline system
Xiaolou Li
  • paper reading
Zehua Liu
  • VTS paper reading and baseline
  • SadTalker begin
Pengqi Li
  • Phone's discrimination in deep SID [3]
Wan Lin
  • NC paper
Tianhao Wang
  • SE fine-tuning [4]
  • VoxTube DataSet
Zhenyu Zhou
Junhui Chen
  • VoxTube DataSet
Jiaying Wang
  • beam search paper reading
  • pure beam search using variance as weight [6]
Yu Zhang
  • Crowd sound event detection test
  • EER test
  • Model size analyze [7]
Wenqiang Du
  • some tests about the project
Yang Wei
  • Prepare FunASR for shengyuanzhineng