“2025-02-10”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（3位用户的4个中间修订版本未显示）
第8行： 第8行：
 
* AI handbook high-school version v3.0 done
 
* AI handbook high-school version v3.0 done
 
* All pictures for handbook done
 
* All pictures for handbook done
−
 
−
 
 
||
 
||
 
*
 
*
第20行： 第18行：
 
|Lantian Li
 
|Lantian Li
 
||
 
||
−
*  
+
* Final version proofreading of the high-school book (3/40)
 +
* Preview IS2025 papers.
 
||
 
||
 
*
 
*
第31行： 第30行：
 
|Ying Shi
 
|Ying Shi
 
||
 
||
−
*  
+
* https://z1et6d3xtb.feishu.cn/docx/Wg3ldbCKeoKhBzxZ2lccTO8ZnTf?from=from_copylink
 
||
 
||
 
*  
 
*  
第54行： 第53行：
 
|Junming Yuan
 
|Junming Yuan
 
||
 
||
−
*  
+
* High-school PPT and Jiaoan(4)
 
||
 
||
 
*
 
*
第201行： 第200行：
 
|Yang Wei
 
|Yang Wei
 
||
 
||
−
*  
+
* Train text enroll kws model with short audio data. (recall mandarin: 80%, yue: 88%)
 +
* Train model with larger model; finetune with fyt recording data. (In progress)
 
||
 
||
 
*
 
*

2025年2月11日 (二) 03:13的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • AI handbook high-school version v3.0 done
  • All pictures for handbook done
Lantian Li
  • Final version proofreading of the high-school book (3/40)
  • Preview IS2025 papers.
Ying Shi
Zhenghai You
  • Revise the Is2025 paper to the second edition
  • Reading some papers on SE/SS/TSE refiner
Junming Yuan
  • High-school PPT and Jiaoan(4)
Xiaolou Li
  • Write CNVSRC2024 paper
  • Config Gongan Server and copy data to it
  • pre-process cvs3 data
Zehua Liu
  • Writing CNVSRC-2024 paper
  • Writing VSR-LLM paper and Doing Experiment
  • AlignVSR Current Result on LRS3 （WER：33.9%） < SyncVSR（WER：33.3%）(Maybe need change hyper-parameter)
Pengqi Li
  • High-school PPT and Jiaoan(5)
Wan Lin
  • Revised NS paper for IS2025 [1] (still lack of experimental results)
  • Train multi-scenario model (need parameters adjusting, still in training)
  • Try other revised-BCE loss(failed)
Tianhao Wang
  • reproducing two baselines: Universal sound separation and MixIT
Xiaoxue Luo
  • Check the pictures in AI high school handbook,completed the first chapter(1/4)
Zhenyu Zhou
  • write graduation paper
  • conditional chain experiment & paper draft with jiaying
Junhui Chen
  • faster test code for NS (Vox-O 25min -> 5min)
  • writing paper for is25
  • (On plane to Beijing)
Jiaying Wang
  • is25 paper: abstract, intro, related work, method, dataset parts have been finished[2]
  • experiment: baseline with pit and cc: training and test all finished
  • model which sort by ctc: coding, finished in these days
Yu Zhang
  • refine prompt
  • add backtesting report to debate report
  • trying stronger model (DeepSeek R1 70B)
  • introduce more trading groups that are mutually opposed.
Wenqiang Du
  • AI primary handbook PPT （16/35）
  • Check AI primary handbook
  • Check AI middle handbook
Yang Wei
  • Train text enroll kws model with short audio data. (recall mandarin: 80%, yue: 88%)
  • Train model with larger model; finetune with fyt recording data. (In progress)
Turi
  • Text data collection
  • Adding Language Model to ASR (not successful yet)
Yue Gu
  • almost finish phone-level gaussian model training
  • according to the badcase analysis，I‘m designing a new framework for fine-grained speaker adaptation
Qi Qu
  • Porting Text-enroll KWS models (delivery 2025-01-08) to mr536 NPU.
取自“http://cslt.org/mediawiki/index.php?title=2025-02-10&oldid=42803”