“2024-09-09”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(9位用户的9个中间修订版本未显示)
第39行: 第39行:
 
|Zhenghai You
 
|Zhenghai You
 
||
 
||
*
+
* ICASSP2025 experiment and writting (Spk-Aug in TSE)
 
||
 
||
 
*
 
*
第51行: 第51行:
 
* mixed Hubert pretraining v1 result[https://z1et6d3xtb.feishu.cn/docx/O7cIdWjYZoVelYxJqqUcMjLmnxf](error)
 
* mixed Hubert pretraining v1 result[https://z1et6d3xtb.feishu.cn/docx/O7cIdWjYZoVelYxJqqUcMjLmnxf](error)
 
** find a aligned problem(get a error aligned masked label)
 
** find a aligned problem(get a error aligned masked label)
||
 
*
 
||
 
*
 
|-
 
 
 
|-
 
|Chen Chen
 
||
 
*
 
 
||
 
||
 
*
 
*
第72行: 第61行:
 
|Xiaolou Li
 
|Xiaolou Li
 
||
 
||
*
+
* IS2024 conference
 +
* organize papers and posters
 
||
 
||
 
*
 
*
第94行: 第84行:
 
|Pengqi Li
 
|Pengqi Li
 
||
 
||
*
+
* Reproduced LayerCAM & TAO on SV[https://z1et6d3xtb.feishu.cn/docx/ZsWRd4qi8ocZEwx6oPbcZG2unvc].
 +
* But two methods consistency is not enough compare with previous work.
 
||
 
||
 
*
 
*
第105行: 第96行:
 
|Wan Lin
 
|Wan Lin
 
||
 
||
*
+
* ICASSP 2025 paper writting
 
||
 
||
 
*
 
*
第116行: 第107行:
 
|Tianhao Wang
 
|Tianhao Wang
 
||
 
||
*
+
* IS2024 conference
 
||
 
||
 
*
 
*
第127行: 第118行:
 
|Zhenyu Zhou
 
|Zhenyu Zhou
 
||
 
||
*
+
*ICASSP 2025 paper writting
 
||
 
||
 
*
 
*
第138行: 第129行:
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
*
+
* ICASSP 2025 paper writting
 
||
 
||
 
*
 
*
第160行: 第151行:
 
|Yu Zhang
 
|Yu Zhang
 
||
 
||
*
+
* Huawei AED acceptance report
 +
* Abuse detection model retrain
 
||
 
||
 
*
 
*
第171行: 第163行:
 
|Wenqiang Du
 
|Wenqiang Du
 
||
 
||
*
+
*Release chinese and Cantonese KWS model
 
||
 
||
 
*
 
*
第182行: 第174行:
 
|Yang Wei
 
|Yang Wei
 
||
 
||
*
+
* Test text enroll KWS model.
 
||
 
||
 
*
 
*
第215行: 第207行:
 
||
 
||
 
* Interspeech2024
 
* Interspeech2024
* read several interspeech2024 papers about contextual ASR
+
* Read several interspeech2024 papers about contextual ASR
 
||
 
||
 
*
 
*
第224行: 第216行:
 
|Qi Qu
 
|Qi Qu
 
||
 
||
*  
+
* KWS:
 +
** Prepared test datasets and located ideal thresholds: Mandarin Chinese 48 words, Cantonese 10 words, and Uyghur 10 words.
 +
** Defined several scenes (different cost coefficients): school, meeting room, demonstration, etc.
 +
** Standardized test routines w/ AIBabel Changzhou team.
 
||
 
||
 
*
 
*

2024年9月9日 (一) 11:06的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Handbook for middle school. v0.0 done.
Lantian Li
Ying Shi
  • INTERSPEECH 2024
  • Help Turi and Zhenghai for the ICASSP paper
Zhenghai You
  • ICASSP2025 experiment and writting (Spk-Aug in TSE)
Junming Yuan
  • mixed Hubert pretraining v1 result[1](error)
    • find a aligned problem(get a error aligned masked label)
Xiaolou Li
  • IS2024 conference
  • organize papers and posters
Zehua Liu
Pengqi Li
  • Reproduced LayerCAM & TAO on SV[2].
  • But two methods consistency is not enough compare with previous work.
Wan Lin
  • ICASSP 2025 paper writting
Tianhao Wang
  • IS2024 conference
Zhenyu Zhou
  • ICASSP 2025 paper writting
Junhui Chen
  • ICASSP 2025 paper writting
Jiaying Wang
Yu Zhang
  • Huawei AED acceptance report
  • Abuse detection model retrain
Wenqiang Du
  • Release chinese and Cantonese KWS model
Yang Wei
  • Test text enroll KWS model.
Lily
  • Record AI lesson (2/10)
    • Chapter 0(introduction)
    • Chapter 1
  • Paper reading (about Chinese Mispronunciation Detection)
  • AI radiance's daily work
Turi
  • Dataset paper writing for icassp
Yue Gu
  • Interspeech2024
  • Read several interspeech2024 papers about contextual ASR
Qi Qu
  • KWS:
    • Prepared test datasets and located ideal thresholds: Mandarin Chinese 48 words, Cantonese 10 words, and Uyghur 10 words.
    • Defined several scenes (different cost coefficients): school, meeting room, demonstration, etc.
    • Standardized test routines w/ AIBabel Changzhou team.