“2024-03-04”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(15位用户的20个中间修订版本未显示)
第6行: 第6行:
 
|Dong Wang
 
|Dong Wang
 
||  
 
||  
*  
+
* Interspeech papers
 +
* ICME review done
 +
* XXX project management
 +
 
 
||
 
||
 
*
 
*
第30行: 第33行:
 
|Ying Shi
 
|Ying Shi
 
||  
 
||  
*
+
* INTERSPEECH Paper [https://z1et6d3xtb.feishu.cn/wiki/E5v0wvIgoiT0ivkXnTdcV9ygnLx?from=from_copylink here]
 
||
 
||
 
*  
 
*  
第51行: 第54行:
 
|Junming Yuan
 
|Junming Yuan
 
||  
 
||  
*
+
* INTERSPEECH 2024 Papers
 +
** the results still have problem in clean scenario (triple check tonight)
 
||
 
||
*
+
* continue to work on INTERSPEECH 2024 papers
 +
** make additional experiment
 
||
 
||
 
*   
 
*   
第74行: 第79行:
 
|Xiaolou Li
 
|Xiaolou Li
 
||  
 
||  
*  
+
* INTERSPEECH 2024 Papers
 
||
 
||
 
*  
 
*  
第85行: 第90行:
 
|Zehua Liu
 
|Zehua Liu
 
||  
 
||  
*
+
*is24 papper writing
 +
 
 
||
 
||
 
*  
 
*  
第96行: 第102行:
 
|Pengqi Li
 
|Pengqi Li
 
||   
 
||   
*  
+
* Team Work[https://z1et6d3xtb.feishu.cn/docx/T3U2dTs5poiIgtxtM2Sc0QennWe]
 +
** Synchronous experiments and hypotheses with Liuhuan about vad-free SID
 +
** JinFu:Update Linear transform for channel mismatch according the last discussion
 +
** A finding from XueYing: pretrianed model by long duration and then finetune by more short duration is work!
 
||
 
||
 
*
 
*
第107行: 第116行:
 
|Wan Lin
 
|Wan Lin
 
||  
 
||  
*  
+
* neural scoring [https://z1et6d3xtb.feishu.cn/docx/TQvWdk8LVo9ONaxQ5Qac9A2Dn3d?from=from_copylink]
 
||
 
||
 
*
 
*
第118行: 第127行:
 
|Tianhao Wang
 
|Tianhao Wang
 
||  
 
||  
*  
+
* INTERSPEECH 2024 Paper
 +
** Close-Set limited data SID testing ...
 +
** paper refinement ing ...
 
||
 
||
 
*  
 
*  
第129行: 第140行:
 
|Zhenyu Zhou
 
|Zhenyu Zhou
 
||  
 
||  
*
+
* INTERSPEECH 2024 Papers
 +
* Detailed comparisons and analyses of SP & VTLP[https://z1et6d3xtb.feishu.cn/docx/VLbOdzlkUochGQxLUiNcg4IznXe]
 
||
 
||
 
*
 
*
第140行: 第152行:
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
*  
+
* Try to modify model structure for Neural Scoring
 
||
 
||
 
*
 
*
第151行: 第163行:
 
|Jiaying Wang
 
|Jiaying Wang
 
||  
 
||  
*  
+
* report on last Friday
 +
* continue to verify the influence of speaker encoder and randomness of cohort
 +
 
 
||
 
||
 
*  
 
*  
第162行: 第176行:
 
|Yu Zhang
 
|Yu Zhang
 
||
 
||
*  
+
* financial-pipeline
 +
** check logic with Jun Wang and debug
 +
** factor analysis visualization
 
||
 
||
*
+
* portfolio analysis visualization
 +
* finish debug
 
||
 
||
 
*   
 
*   
第173行: 第190行:
 
|Wenqiang Du
 
|Wenqiang Du
 
||  
 
||  
*  
+
* Aibabel
 +
**Actual scene noise (toilet, dormitory) collection
 +
**KWS model for CN Uyghur  and  train and cn and Uyghur Joint Model  train and test(not compeleted)[https://z1et6d3xtb.feishu.cn/docx/S67Dd7TVNoNKEaxELiXc7qE6nrd?from=from_copylink]
 
||
 
||
 
*
 
*
第184行: 第203行:
 
|Yang Wei
 
|Yang Wei
 
||  
 
||  
*  
+
* Rearrange some customer project directory and prepare the release directory.
 
||
 
||
 
*
 
*
第194行: 第213行:
 
|Lily
 
|Lily
 
||
 
||
*  
+
* Paper reading
 +
* Prepare for overview paper
 
||
 
||
 
*
 
*
第204行: 第224行:
 
|Turi
 
|Turi
 
||
 
||
*  
+
* Read KWS Papers
 +
* Experiment with Few-Shot KWS on Oromo language
 
||
 
||
 
*
 
*

2024年3月4日 (一) 11:35的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Interspeech papers
  • ICME review done
  • XXX project management
Lantian Li
  • GPU status [1]
  • ASIP-BUPT (CohortTSE, SE-Adapter, SpeakerAug, NeuralScoring)
  • INTERSPEECH 2024
Ying Shi
  • INTERSPEECH Paper here
Zhenghai You
Junming Yuan
  • INTERSPEECH 2024 Papers
    • the results still have problem in clean scenario (triple check tonight)
  • continue to work on INTERSPEECH 2024 papers
    • make additional experiment
Chen Chen
  • INTERSPEECH 2024 Papers
  • System Fusion for DeepFake Detection [2] GOOD RESULT!
  • continue to work on INTERSPEECH 2024 papers
Xiaolou Li
  • INTERSPEECH 2024 Papers
Zehua Liu
  • is24 papper writing
Pengqi Li
  • Team Work[3]
    • Synchronous experiments and hypotheses with Liuhuan about vad-free SID
    • JinFu:Update Linear transform for channel mismatch according the last discussion
    • A finding from XueYing: pretrianed model by long duration and then finetune by more short duration is work!
Wan Lin
  • neural scoring [4]
Tianhao Wang
  • INTERSPEECH 2024 Paper
    • Close-Set limited data SID testing ...
    • paper refinement ing ...
Zhenyu Zhou
  • INTERSPEECH 2024 Papers
  • Detailed comparisons and analyses of SP & VTLP[5]
Junhui Chen
  • Try to modify model structure for Neural Scoring
Jiaying Wang
  • report on last Friday
  • continue to verify the influence of speaker encoder and randomness of cohort
Yu Zhang
  • financial-pipeline
    • check logic with Jun Wang and debug
    • factor analysis visualization
  • portfolio analysis visualization
  • finish debug
Wenqiang Du
  • Aibabel
    • Actual scene noise (toilet, dormitory) collection
    • KWS model for CN Uyghur and train and cn and Uyghur Joint Model train and test(not compeleted)[6]
Yang Wei
  • Rearrange some customer project directory and prepare the release directory.
Lily
  • Paper reading
  • Prepare for overview paper
Turi
  • Read KWS Papers
  • Experiment with Few-Shot KWS on Oromo language