“2024-10-14”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（16位用户的23个中间修订版本未显示）
第6行： 第6行：
 
|Dong Wang
 
|Dong Wang
 
||
 
||
−
*
+
* AI handbook high-education version, experiment booklet
 +
* Check AI primary school handbook (1-20)
 
||
 
||
 
*
 
*
第17行： 第18行：
 
|Lantian Li
 
|Lantian Li
 
||
 
||
−
*
+
* AI-Graph EN (20/50)
 +
* Prepare CSTR intro report
 
||
 
||
 
*
 
*
第28行： 第30行：
 
|Ying Shi
 
|Ying Shi
 
||
 
||
−
*  
+
* Finish Text enroll keywords spotting code & document and deliver to Wei & Du
 +
* Cohort Overlap ASR code v0.0
 +
** code has finished and training has been done
 +
* Cohort Speech separation code v0.0
 +
** code has finished training is in progress
 +
* [https://z1et6d3xtb.feishu.cn/docx/OHjsdgVmhoXUGpxvh5tcaBN4nAh?from=from_copylink here]
 
||
 
||
 
*
 
*
第39行： 第46行：
 
|Zhenghai You
 
|Zhenghai You
 
||
 
||
−
*
+
* Exploring the role of speaker encoder in TSE and generality of SPK-AUG[https://z1et6d3xtb.feishu.cn/docx/GHF8doRjDo50ihxGUPpcsZgLncb?from=space_home_recent&pre_pathname=%2Fdrive%2Fhome%2F&previous_navigation_time=1728902573829]
 
||
 
||
 
*
 
*
第49行： 第56行：
 
|Junming Yuan
 
|Junming Yuan
 
||
 
||
−
*
+
* MT-Hubert exp[https://z1et6d3xtb.feishu.cn/docx/IXBzdx2OvoDbvLxBLmKcjGJZnvb]:
 +
** codebook set + infoNCE ---> FC+softmax+CE / FC+sigmoid+BCE
 +
*** To reduce the learning rate can work.
 +
** verified the feat-mask MT-Hubert with different lr
 +
** time-mask MT-Hubert verification (in progress)
 
||
 
||
 
*
 
*
第71行： 第82行：
 
|Xiaolou Li
 
|Xiaolou Li
 
||
 
||
−
*
+
* AV-HuBERT discrete unit training (wer: ↓1.5-3%)
 +
** rethink how to prove the advantage or disadvantage of discrete unit?
 +
* Dense connector experiments (in training)
 +
* Double check the data of existing 3000h data in CVS2
 +
* Paper reading (discrete unit, VTS)
 
||
 
||
−
*
+
* Design a experiment to explain the performance of discrete unit
 +
* Finish data double check
 +
* Try to establish a simple VTS system based on our VSR system
 
||
 
||
 
*
 
*
第82行： 第99行：
 
|Zehua Liu
 
|Zehua Liu
 
||
 
||
−
*
+
*Av-Hubert(Frozen) as Encoder performe very bad(cer:80%)[https://z1et6d3xtb.feishu.cn/docx/JBsidACDVojhCaxFQLbcCVbsnAc?from=from_copylink]
 +
**after finetune maybe better ,but still bad
 +
*Qwen-14B perform better(47%) than Qwen-7B(50%)
 +
*Finish In-Context-Learning code and is training
 +
** maybe i will get result very soon
 
||
 
||
−
*
+
*verify collected data with XiaoLou
 +
*finish VTS data Acceptance report
 
||
 
||
 
*
 
*
第93行： 第115行：
 
|Pengqi Li
 
|Pengqi Li
 
||
 
||
−
*
+
* Evaluate TAO and LayerCAM(verification) reliability.
 +
** Exploring the Consistency of TAO and LayerCAM Results on different models and datasets.
 
||
 
||
 
*
 
*
第104行： 第127行：
 
|Wan Lin
 
|Wan Lin
 
||
 
||
−
*
+
* NS
 +
** poster
 +
** data preparing and processing
 +
** adjust the training code
 
||
 
||
 
*
 
*
第115行： 第141行：
 
|Tianhao Wang
 
|Tianhao Wang
 
||
 
||
−
*
+
* CLIPSep exps for 2-mix and 5-mix [https://z1et6d3xtb.feishu.cn/docx/DnJgdwtNhotEpIxH7zfcksETnte]
 +
** 2-mix(whole vggsound, 300 classes): SDR-mix = -1.1748, SDR-separate = 5.0145
 +
** 5-mix(50 classes of vggsound): SDR-mix = -11.4529, SDR-separate = -0.4764
 
||
 
||
 
*
 
*
第139行： 第167行：
 
|Zhenyu Zhou
 
|Zhenyu Zhou
 
||
 
||
−
*
+
*Model quantization version2
 +
*Multi-talker mix data preparation
 
||
 
||
 
*
 
*
第150行： 第179行：
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
−
*
+
* Prepare vb2 data
 +
** Too many utterances for training (out of memory), thinking a smart way to divide them.
 
||
 
||
 
*
 
*
第172行： 第202行：
 
|Yu Zhang
 
|Yu Zhang
 
||
 
||
−
*
+
* SocioDojo Llama version
 +
** news integration is adjusted once every 12 hours
 +
** wikipedia & google search is banned
 
||
 
||
 
*
 
*
第183行： 第215行：
 
|Wenqiang Du
 
|Wenqiang Du
 
||
 
||
−
*
+
*Check the data from past training models and update the KWS model again（Model testing）
 +
** Chinese, Cantonese, Minnan, Haining and Uyghur
 
||
 
||
 
*
 
*
第194行： 第227行：
 
|Yang Wei
 
|Yang Wei
 
||
 
||
−
*
+
* Train text enroll KWS model with updated code (in progress)
 
||
 
||
 
*
 
*
第214行： 第247行：
 
|Turi
 
|Turi
 
||
 
||
−
*
+
* Whisper model finetuning[https://uestc.feishu.cn/docx/LsVmd6Mr3o90U2xVULOcImPmnjb?from=from_copylink]
 
||
 
||
 
*
 
*
第222行： 第255行：
 
|Yue Gu
 
|Yue Gu
 
||
 
||
−
*
+
* revise the TASLP paper
 +
* read several papers about accent and prosody
 
||
 
||
 
*
 
*
第231行： 第265行：
 
|Qi Qu
 
|Qi Qu
 
||
 
||
−
*  
+
* AED: classifiers retrained w/ new method (suppression on negative stimuli) and improvement attested.
 
||
 
||
 
*
 
*

2024年10月14日 (一) 11:02的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • AI handbook high-education version, experiment booklet
  • Check AI primary school handbook (1-20)
Lantian Li
  • AI-Graph EN (20/50)
  • Prepare CSTR intro report
Ying Shi
  • Finish Text enroll keywords spotting code & document and deliver to Wei & Du
  • Cohort Overlap ASR code v0.0
    • code has finished and training has been done
  • Cohort Speech separation code v0.0
    • code has finished training is in progress
  • here
Zhenghai You
  • Exploring the role of speaker encoder in TSE and generality of SPK-AUG[1]
Junming Yuan
  • MT-Hubert exp[2]:
    • codebook set + infoNCE ---> FC+softmax+CE / FC+sigmoid+BCE
      • To reduce the learning rate can work.
    • verified the feat-mask MT-Hubert with different lr
    • time-mask MT-Hubert verification (in progress)
Chen Chen
Xiaolou Li
  • AV-HuBERT discrete unit training (wer: ↓1.5-3%)
    • rethink how to prove the advantage or disadvantage of discrete unit?
  • Dense connector experiments (in training)
  • Double check the data of existing 3000h data in CVS2
  • Paper reading (discrete unit, VTS)
  • Design a experiment to explain the performance of discrete unit
  • Finish data double check
  • Try to establish a simple VTS system based on our VSR system
Zehua Liu
  • Av-Hubert(Frozen) as Encoder performe very bad(cer:80%)[3]
    • after finetune maybe better ,but still bad
  • Qwen-14B perform better(47%) than Qwen-7B(50%)
  • Finish In-Context-Learning code and is training
    • maybe i will get result very soon
  • verify collected data with XiaoLou
  • finish VTS data Acceptance report
Pengqi Li
  • Evaluate TAO and LayerCAM(verification) reliability.
    • Exploring the Consistency of TAO and LayerCAM Results on different models and datasets.
Wan Lin
  • NS
    • poster
    • data preparing and processing
    • adjust the training code
Tianhao Wang
  • CLIPSep exps for 2-mix and 5-mix [4]
    • 2-mix(whole vggsound, 300 classes): SDR-mix = -1.1748, SDR-separate = 5.0145
    • 5-mix(50 classes of vggsound): SDR-mix = -11.4529, SDR-separate = -0.4764
Xiaoxue Luo
  • Paper reading about sound separation
  • AudioSep reproduction
    • Training time is too long -> replace with a small dataset(in training)
Zhenyu Zhou
  • Model quantization version2
  • Multi-talker mix data preparation
Junhui Chen
  • Prepare vb2 data
    • Too many utterances for training (out of memory), thinking a smart way to divide them.
Jiaying Wang
Yu Zhang
  • SocioDojo Llama version
    • news integration is adjusted once every 12 hours
    • wikipedia & google search is banned
Wenqiang Du
  • Check the data from past training models and update the KWS model again（Model testing）
    • Chinese, Cantonese, Minnan, Haining and Uyghur
Yang Wei
  • Train text enroll KWS model with updated code (in progress)
Lily
Turi
  • Whisper model finetuning[5]
Yue Gu
  • revise the TASLP paper
  • read several papers about accent and prosody
Qi Qu
  • AED: classifiers retrained w/ new method (suppression on negative stimuli) and improvement attested.
取自“http://index.cslt.org/mediawiki/index.php?title=2024-10-14&oldid=42317”