“2024-06-03”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（18位用户的26个中间修订版本未显示）
第6行： 第6行：
 
|Dong Wang
 
|Dong Wang
 
||
 
||
−
*
+
* Paper for Uyghur children pronunciation analysis (@Maolidan)
 +
* Review NSFC
 +
* Several public talks
 
||
 
||
 
*
 
*
第17行： 第19行：
 
|Lantian Li
 
|Lantian Li
 
||
 
||
−
*
+
* GPU status [https://z1et6d3xtb.feishu.cn/wiki/XGcGwRK5viJmpRkjH9AczIhynCh]
 +
* Projects
 +
** AED -> Quantization
 +
** TSE -> SpkAug, InstanceNorm, AbsentLoss
 +
** VSR -> reaching 1000h
 +
** Finance -> Few-shot, RAG + LLM
 +
* Papers
 +
** NeuralScoring -> Paper-Exps, Speed up the progress
 +
* Postgraduate thesis & defense
 
||
 
||
 
*
 
*
第28行： 第38行：
 
|Ying Shi
 
|Ying Shi
 
||
 
||
−
*
+
* Finish phone-level CTC decoding with TLG fst
 +
* Some stuff about Huawei project collaboration
 
||
 
||
−
*
+
* Restart research about Cohort Overlap ASR
 
||
 
||
 
*
 
*
第39行： 第50行：
 
|Zhenghai You
 
|Zhenghai You
 
||
 
||
−
*
+
* Revised dataset with correct snr
 +
* Train targetless data with SE-SISDR loss[https://z1et6d3xtb.feishu.cn/docx/ByKfdNBcLolQsWxMl3hcjI0XnHh]
 
||
 
||
 
*
 
*
第49行： 第61行：
 
|Junming Yuan
 
|Junming Yuan
 
||
 
||
−
*
+
* Adjusted the strategy of "learn not to listen", but still no work[https://z1et6d3xtb.feishu.cn/docx/Ua0cdv3ano0qHoxN8YvcmRsVn9f]
 +
* make the plan for the next stage
 
||
 
||
 
*
 
*
第71行： 第84行：
 
|Xiaolou Li
 
|Xiaolou Li
 
||
 
||
−
*
+
* Pos embedding Exp [https://z1et6d3xtb.feishu.cn/docx/MjMpdxyjAoK5I7xuwThcqdfkngd?from=from_copylink]
 +
* Frontend and bi-Transformer Exp
 +
* Branchmamba Exp
 
||
 
||
 
*
 
*
第82行： 第97行：
 
|Zehua Liu
 
|Zehua Liu
 
||
 
||
−
*
+
*Smooth Label Celoss (cer: 48.09)[https://z1et6d3xtb.feishu.cn/docx/ZaTFd3A5EoK982xWBVschloanee?from=from_copylink]
 +
*Find some parameters need to adjust[https://z1et6d3xtb.feishu.cn/docx/I8nKdBPz1ojcskxNbDocvvzenud?from=from_copylink]
 
||
 
||
 
*
 
*
第93行： 第109行：
 
|Pengqi Li
 
|Pengqi Li
 
||
 
||
−
*
+
* Completed the NC paper(v1) titled "Explainable Artificial Intelligence for Deep Speech Signal Processing: A survey"
 
||
 
||
 
*
 
*
第104行： 第120行：
 
|Wan Lin
 
|Wan Lin
 
||
 
||
−
*
+
* Supplement experimental results
 +
** NS & Mixup & Eaasp single/mutil-scenario training
 
||
 
||
 
*
 
*
第115行： 第132行：
 
|Tianhao Wang
 
|Tianhao Wang
 
||
 
||
−
*
+
* Tracing experimental results
 +
** "hard spk sample" is useful for clean ns
 +
* share encoder exps (training)
 
||
 
||
 
*
 
*
第126行： 第145行：
 
|Zhenyu Zhou
 
|Zhenyu Zhou
 
||
 
||
−
*
+
*huawei project[https://z1et6d3xtb.feishu.cn/docx/ByKfdNBcLolQsWxMl3hcjI0XnHh]
 
||
 
||
 
*
 
*
第137行： 第156行：
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
−
*
+
* Neural Scoring complementary experiments
 
||
 
||
 
*
 
*
第148行： 第167行：
 
|Jiaying Wang
 
|Jiaying Wang
 
||
 
||
−
*
+
* come back! continue cohort within vit structure
 
||
 
||
 
*
 
*
第159行： 第178行：
 
|Yu Zhang
 
|Yu Zhang
 
||
 
||
−
*
+
* Paper rading
 +
* Huawei cough humming model quantization
 
||
 
||
 
*
 
*
第170行： 第190行：
 
|Wenqiang Du
 
|Wenqiang Du
 
||
 
||
−
*
+
* Prepare to train new kws model （48 words） for Aibabel Company
 
||
 
||
 
*
 
*
第181行： 第201行：
 
|Yang Wei
 
|Yang Wei
 
||
 
||
−
*
+
* Huilan TTS
 +
** Fix bugs of the updated TTS server.
 +
** Try to use ONNX model in TTS server. (In progress)
 
||
 
||
 
*
 
*
第191行： 第213行：
 
|Lily
 
|Lily
 
||
 
||
−
*
+
* Live broadcast
 +
* Xinjiang teacher's course
 +
* Children's course (Sunday)
 
||
 
||
 
*
 
*
第201行： 第225行：
 
|Turi
 
|Turi
 
||
 
||
−
*
+
* Data collection
 +
** No contribution, as students are on final
 +
* Final projects and live broadcast
 
||
 
||
 
*
 
*
第209行： 第235行：
 
|Yue Gu
 
|Yue Gu
 
||
 
||
−
*
+
* complete the semantic paraformer exps
 +
* do some modular analysis about Competition Elimination Mechanism
 
||
 
||
 
*
 
*
第218行： 第245行：
 
|Qi Qu
 
|Qi Qu
 
||
 
||
−
*  
+
* KWS
 +
** Data processing routines: VAD, ASR.
 +
** Test routines: keyword-wise ROC.
 +
** Data collection: Uyghur and Kazakh.
 +
* AED
 +
** Online demo.
 
||
 
||
 
*
 
*

2024年6月3日 (一) 11:02的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Paper for Uyghur children pronunciation analysis (@Maolidan)
  • Review NSFC
  • Several public talks
Lantian Li
  • GPU status [1]
  • Projects
    • AED -> Quantization
    • TSE -> SpkAug, InstanceNorm, AbsentLoss
    • VSR -> reaching 1000h
    • Finance -> Few-shot, RAG + LLM
  • Papers
    • NeuralScoring -> Paper-Exps, Speed up the progress
  • Postgraduate thesis & defense
Ying Shi
  • Finish phone-level CTC decoding with TLG fst
  • Some stuff about Huawei project collaboration
  • Restart research about Cohort Overlap ASR
Zhenghai You
  • Revised dataset with correct snr
  • Train targetless data with SE-SISDR loss[2]
Junming Yuan
  • Adjusted the strategy of "learn not to listen", but still no work[3]
  • make the plan for the next stage
Chen Chen
Xiaolou Li
  • Pos embedding Exp [4]
  • Frontend and bi-Transformer Exp
  • Branchmamba Exp
Zehua Liu
  • Smooth Label Celoss (cer: 48.09)[5]
  • Find some parameters need to adjust[6]
Pengqi Li
  • Completed the NC paper(v1) titled "Explainable Artificial Intelligence for Deep Speech Signal Processing: A survey"
Wan Lin
  • Supplement experimental results
    • NS & Mixup & Eaasp single/mutil-scenario training
Tianhao Wang
  • Tracing experimental results
    • "hard spk sample" is useful for clean ns
  • share encoder exps (training)
Zhenyu Zhou
  • huawei project[7]
Junhui Chen
  • Neural Scoring complementary experiments
Jiaying Wang
  • come back! continue cohort within vit structure
Yu Zhang
  • Paper rading
  • Huawei cough humming model quantization
Wenqiang Du
  • Prepare to train new kws model （48 words） for Aibabel Company
Yang Wei
  • Huilan TTS
    • Fix bugs of the updated TTS server.
    • Try to use ONNX model in TTS server. (In progress)
Lily
  • Live broadcast
  • Xinjiang teacher's course
  • Children's course (Sunday)
Turi
  • Data collection
    • No contribution, as students are on final
  • Final projects and live broadcast
Yue Gu
  • complete the semantic paraformer exps
  • do some modular analysis about Competition Elimination Mechanism
Qi Qu
  • KWS
    • Data processing routines: VAD, ASR.
    • Test routines: keyword-wise ROC.
    • Data collection: Uyghur and Kazakh.
  • AED
    • Online demo.
取自“http://index.cslt.org/mediawiki/index.php?title=2024-06-03&oldid=41838”