“2024-05-27”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
(10位用户的12个中间修订版本未显示)
第6行: 第6行:
 
|Dong Wang
 
|Dong Wang
 
||
 
||
*
+
* Several public talk (Henan, Aizhe, ChongQing, Kexie, Huawei)
 +
* NSFC review
 +
* Eletronics review
 +
 
 
||
 
||
 
*
 
*
第63行: 第66行:
 
|Chen Chen
 
|Chen Chen
 
||
 
||
*
+
* thesis defence
 +
* vii group
 +
** Structure: some experiment about 3d-cnn frontend with mamba on LRS2
 +
** Strategy: more experiments on audio memory (need adjust some hyper-parameters)
 +
** Data: 73 hours new, 612 hours in total
 
||
 
||
 
*
 
*
第85行: 第92行:
 
|Zehua Liu
 
|Zehua Liu
 
||
 
||
*
+
*AKVSR+pos_emb+attscore_CEloss(48.40) > AKVSR+pos_emb(48.80)[https://z1et6d3xtb.feishu.cn/docx/ZaTFd3A5EoK982xWBVschloanee?from=from_copylink]
 +
*NCMMSC2024 papper finish and i need check again
 
||
 
||
 
*
 
*
第96行: 第104行:
 
|Pengqi Li
 
|Pengqi Li
 
||
 
||
*
+
* NC Papers
 +
* live broadcast
 
||
 
||
 
*
 
*
第107行: 第116行:
 
|Wan Lin
 
|Wan Lin
 
||
 
||
*
+
* NS: Complementary experiments
 
||
 
||
 
*
 
*
第118行: 第127行:
 
|Tianhao Wang
 
|Tianhao Wang
 
||
 
||
*
+
* Tracing experimental results
 +
* The influence of overlap ratio on EER threshold [https://z1et6d3xtb.feishu.cn/docx/BywjdkGvNou12sxQ4dAcxYa9noh]
 
||
 
||
 
*
 
*
第140行: 第150行:
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
*
+
* Neural Scoring complementary experiments
 
||
 
||
 
*
 
*
第184行: 第194行:
 
|Yang Wei
 
|Yang Wei
 
||
 
||
*
+
* Huilan stuff
 +
** Update TTS server for multiprocessing, to reduce latency.
 
||
 
||
 
*
 
*
第204行: 第215行:
 
|Turi
 
|Turi
 
||
 
||
*
+
* Data collection
 +
** Audio checking for correctness, 26.5K so far.
 +
* Course works
 
||
 
||
 
*
 
*
第212行: 第225行:
 
|Yue Gu
 
|Yue Gu
 
||
 
||
*
+
* write paper: exp part 100%, method part 20%
 +
* complete Kespeech exps, performance meets expectations
 +
* find a new bug about semantic paraformer, retrain the model
 
||
 
||
 
*
 
*
第221行: 第236行:
 
|Qi Qu
 
|Qi Qu
 
||
 
||
*  
+
* KWS
 +
** (ongoing) Standardize dataset formats and test routines.
 +
** Uyghur dataset: 129 speakers, 10 words.
 +
** New B6-based online model service.
 +
** TUI app for visualization.
 
||
 
||
 
*
 
*

2024年5月27日 (一) 12:04的版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Several public talk (Henan, Aizhe, ChongQing, Kexie, Huawei)
  • NSFC review
  • Eletronics review
Lantian Li
Ying Shi
  • write CTC + fst decoding engine
    • CTC + fst TLG.fst decoding down
    • CTC + fst PLG.fst decoding in progress
Zhenghai You
Junming Yuan
  • make our Vlog and live broadcast video
  • prepare new research plan
Chen Chen
  • thesis defence
  • vii group
    • Structure: some experiment about 3d-cnn frontend with mamba on LRS2
    • Strategy: more experiments on audio memory (need adjust some hyper-parameters)
    • Data: 73 hours new, 612 hours in total
Xiaolou Li
Zehua Liu
  • AKVSR+pos_emb+attscore_CEloss(48.40) > AKVSR+pos_emb(48.80)[1]
  • NCMMSC2024 papper finish and i need check again
Pengqi Li
  • NC Papers
  • live broadcast
Wan Lin
  • NS: Complementary experiments
Tianhao Wang
  • Tracing experimental results
  • The influence of overlap ratio on EER threshold [2]
Zhenyu Zhou
Junhui Chen
  • Neural Scoring complementary experiments
Jiaying Wang
Yu Zhang
  • Paper reading for spatiotemporal time series forecasting and Financial related machine learning
Wenqiang Du
Yang Wei
  • Huilan stuff
    • Update TTS server for multiprocessing, to reduce latency.
Lily
Turi
  • Data collection
    • Audio checking for correctness, 26.5K so far.
  • Course works
Yue Gu
  • write paper: exp part 100%, method part 20%
  • complete Kespeech exps, performance meets expectations
  • find a new bug about semantic paraformer, retrain the model
Qi Qu
  • KWS
    • (ongoing) Standardize dataset formats and test routines.
    • Uyghur dataset: 129 speakers, 10 words.
    • New B6-based online model service.
    • TUI app for visualization.