cslt Wiki - 用户贡献 [zh-cn]

2025-05-26

2025-05-26T10:52:39Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Two days of teaching, two days of training
* Refine NS paper
* College version of the AI hand book
||
*
||
*
|-

|-
|Lantian Li
||
* BUPT Graduate matters
* Huawei projects (N&F)
* Kicking off the translation of ML book
||
*
||
*
|-

|-
|Ying Shi
||
* Prepare some material about the Huang Danian Challenge final report.
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Complete the final delivery of Huawei TSE
||
*
||
*
|-

|-
|Junming Yuan
||
* Find bug and revise my ASR code.
* Review Middle-school AI practice handbook(4).
||
*
||
*
|-

|-
|Xiaolou Li
||
* Continue to audit data
* Reading some paper about SSL
||
* Finish data audition, and start to organize all 10k h data
* Pre on this Friday
||
*
|-

|-
|Zehua Liu
||
*Data Audit and paper revise
*Reading paper about Qwen2.5-omni and maybe apply VTS to this model
*Data augmentation for AlignVSR
||
*
||
*
|-

|-
|Pengqi Li
||
* Research progress: A large number of experiments and visualizations, but there is still confusion in "How to display the explanation".
* Teaching work: Prepare for the AI course at Tsinghua middle School and assist with the graduation ceremony.
||
* Summarize the work that has been completed and review similar papers.
||
*
|-

|-
|Wan Lin
||
* Paper reading, Friday report
* Doing some test
* Revised paper
||
*
||
*
|-

|-
|Tianhao Wang
||
* leave absence
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* prepare Huawei project meeting with tianhao
* the English version of the AI book
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* reading paper
* check NS code, modify and supplement exp results
||
*
||
*
|-

|-
|Jiaying Wang
||
* train pure asr system use clean wsj0
** loss setting and data: remove zero_infinity parameter, use speech longer than 6s
** training: loss can converge at under 20(reasonable), LER 45%(higher than expectation)
** plan: change cnn settings, explore whether the performance is related to word length
* HongYan competition slides modification

||
*
||
*
|-

|-
|Yu Zhang
||
* Profit analysis about Multi Agent System
* Reflection time range experiment
* Some Data analysis task in RoyalFlush (external)
||
*
||
*
|-

|-
|Wenqiang Du
||
* New KWS model for AIbabel(30 words and speech enroll)
||
*
||
*
|-

|-
|Yang Wei
||
* Finish AIbabel online KWS model encryption.
||
*
||
*
|-

|-
|Turi
||
* Modified and submitted the final thesis.
||
*
||
*
|-
|Yue Gu
||
* complete the introduction, method and experimental setup, result is in progress
||
*
||
*
|-
|-
|Qi Qu
||
* Text-enroll KWS models routine work: finding ideal thresholds (word list = Mandarin Chinese 200 words, models = [dist-20250310, dist-20250520]).
||
*
||
*
|-

2025-05-19

2025-05-19T10:40:21Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Training for Tsinghua Primary School
* Training for ZhongLv group
* Refine AIGE book for college students

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* Extended ASR finetuning exp[https://z1et6d3xtb.feishu.cn/docx/Azkwd9VrVosk26xtoZ8cbBnuntc]
* Review Middle-school AI practice handbook(9/20).
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* reading paper
* weekly report
||
*
||
*
|-

|-
|Jiaying Wang
||
* ASR training
** clean data ctc loss converge at 200 (higher than expactation)
** LER have abnormal result：predicted labels tend to concentrate on a few letters, debuging
* Modify the PPT for the hongyan campus competition.
||
*
||
*
|-

|-
|Yu Zhang
||
* For news and sentiment analysis, use reflection as "condition" instead of as "reference" [https://z1et6d3xtb.feishu.cn/docx/SZL2d3y3RolnXlxF02fcK03Bnqd?from=from_copylink]
* Increase news sources (the abuse of reflection is often found when there is a lack of useful news on that day)
* Adding more benchmark (implementing rule based benchmark in TradingAgents (MACD RSI ... ))
||
* Detail Analysis of current backtesting result and check the paper plan with Teacher Li

|-
|Wenqiang Du
||
* Business trip to Changzhou
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Defense
||
*
||
*
|-
|Yue Gu
||
* fix a bug about a reproduced method and retrain the model
* polish my introduction according to some suggestions
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-05-12

2025-05-12T10:54:40Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Training at Tsinghua Primary School
* Training for Zhong Lv group
* Training for Beijing 4th Middle School
* AI education material for college school

||
*
||
*
|-

|-
|Lantian Li
||
* Tackle the Huawei Spark Award
* Review graduate theses
* Engage in several academic visits
||
*
||
*
|-

|-
|Ying Shi
||
* Huawei Spark Award
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Completed the delivery of Huawei TSE project
||
*
||
*
|-

|-
|Junming Yuan
||
* Revise MT-HuBERT paper v0.2(done)
* Need extend ASR finetuning exp:
** clean finetuning on LS100: LER-->4.17% on test-clean
** mix finetuning is in progress.
||
*
||
*
|-

|-
|Xiaolou Li
||
* Added new function to VTS Web, enabling ACC2Speech model choosing
* Polished the presentation slide
* Audited a part of our data between 1 January to 7 April
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||

* Review my research (including related papers, code, etc.).
* Organize a pre-experiment to explore the following questions:
** How to visualize
** How to evaluate reliability
||
* Outline of the paper and pre-experiment
||
*
|-

|-
|Wan Lin
||
* Writing paper, finish 1st version
* Multi-scaled feature exp: training, haven't observed improvement by far
||
*
||
*
|-

|-
|Tianhao Wang
||
* Huawei huohua award delivery
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*read papers and compare the metrics in the audio generation task to select suitable ones to add to our evaluation metrics
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Writing neural scoring paper
* Reading paper
||
*
||
*
|-

|-
|Jiaying Wang
||
* Loss Nan fixed, reason is dimension of linear projection after ctc embedding mismatch the dimension of text label
** after fixed, both linear encoder and conv2d encoder converge well, still training

* Network+ preparation
||
*
||
*
|-

|-
|Yu Zhang
||
* Optimizing Reflection Mechanism (mainly on Rational and The Effective Range of the reflection)
* AED Huawei additional dataset test report delivery (validation model can also reduce the number of parameters on the Huawei dataset)
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check jiaoyubu's AI lesson
* Some of the company's work
** Private deployment of KWS service
** Iteratively upgrade the KWS model using the online data
||
*
||
*
|-

|-
|Yang Wei
||
* Add license check for AIbabel online ASR and KWS (in progress)
||
*
||
*
|-

|-
|Turi
||
* Graduation documents submission
* Preparing for defense
||
*
||
*
|-
|Yue Gu
||
* complete the introduction and related work， method is still in progress（50%）
||
*
||
*
|-
|-
|Qi Qu
||
* Fixed several bugs in converting text-enroll KWS models.
* Some experiments and reports on text-enroll KWS models: [https://b30lttjm7l.feishu.cn/docx/IZlbdBbKpoh5BTxGtlecnMjxnih?from=from_copylink], [https://b30lttjm7l.feishu.cn/docx/Vkjtd1cjboDD3Ex1xMGcq38WnQe?from=from_copylink].
* Consolidating `aibabel_utils` library.
||
*
||
*
|-

2025-04-28

2025-04-28T10:54:36Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
* CSTR report on TSE and USS.
* Dissertation reviews
||
*
||
*
|-

|-
|Ying Shi
||
* thesis
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Some tests of Huawei project
* Writing the report of Huawei project
||
*
||
*
|-

|-
|Junming Yuan
||
* Make Jiaoyubu's AI lesson PPTs with Qiang bro
* Add 3-mixed test results under 2-mix finetuning setting and revise CN version of MT-HuBERT paper
||
*
||
*
|-

|-
|Xiaolou Li
||
* Write GA doc (test report and mid term report)
* Prepare the presentation PPT
* Improve the VTS demo system
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Make Jiaoyubu's AI lesson PPTs with Qiang bro
* Review my research.
||
*
||
*
|-

|-
|Wan Lin
||
* Finish model training, only some test tasks left
* Write paper: finish intro, related, method
||
*
||
*
|-

|-
|Tianhao Wang
||
* Huawei huohua model training
* demo
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* compared CED topk labels with ground truth labels, most CED labels are parent labels of ground truth(eg.ground truth: Male speech,man speaking; CED label: Speech)
* adjust the order of separating audio from low to high probability, it means that our method is effective(SDR: our method: 12.2182; reversed: 11.8793)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Continue to write NS paper (soon discuss with LW)
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* add reflection and memory mechanism based on market feedback (still fine-tuning prompt)
* try other policy fusion method, no significant improvement.
||
*
|-

|-
|Wenqiang Du
||
* Jiaoyubu's AI lesson（10/10）with Junming and Pengqi
* Some tasks about company
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Filled some thesis forms and paper reading
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-04-21

2025-04-21T10:46:26Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Refine the framework of AI text book of primary school.
* Two talks for Tencent.

||
*
||
*
|-

|-
|Lantian Li
||
* Refresh audio retrieval proposal
* Translate ML book
||
*
||
*
|-

|-
|Ying Shi
||
* thesis
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Weekly report
* Huawei project (add cosine loss, get slightly better results)[https://z1et6d3xtb.feishu.cn/wiki/OXubwl2fIip91vkYsgMc1duhnLd]
||
*
||
*
|-

|-
|Junming Yuan
||
* Make Jiaoyubu's AI lesson PPTs with Qiang bro(3)
||
*
||
*
|-

|-
|Xiaolou Li
||
*Gongan Document Writing
*Gongan PPT making
*Making presentation system
||
*
||
*
|-

|-
|Zehua Liu
||
*Gongan Document Writing
*Gongan PPT making
*Making presentation system
||
*
||
*
|-

|-
|Pengqi Li
||
* developed Jiaoyubu's AI lesson PPTs with Qiang bro
||
*
||
*
|-

|-
|Wan Lin
||
* Supply experiments, almost finish, nothing strange
* Start writing paper this week
||
*
||
*
|-

|-
|Tianhao Wang
||
* huawei huohua 2-6mix model training (almost done)
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Exp for NS
* Start writing paper this week
* Reading paper
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Jiaoyubu's AI lesson（7/10）with Junming and Pengqi
||
*
||
*
|-

|-
|Yang Wei
||
* Organizing experiment results for text enroll kws.
||
*
||
*
|-

|-
|Turi
||
* shared conference experience
* Some thesis forms
||
*
||
*
|-
|Yue Gu
||
* write the draft of introduction, complement some exps, draw method figure
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-03-31

2025-03-31T10:56:18Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook college version

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Huawei project cooperation
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Training second iteration TSE model for Huawei project ( eol 0db SI-SDR （once:6.963）-> (twice:8.428))[https://z1et6d3xtb.feishu.cn/wiki/OXubwl2fIip91vkYsgMc1duhnLd]
* Check the breakdown of the host to China Unicom
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-HuBERT paper-CN-v0.1(Done)
* some AI handbook works(checking & revisions) with Brother Qiang
* some AI practice handbook works
* Review primary school PPT
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Paper Reading on last Friday
*Reproducing Accurately Lip2Speech paper，maybe get some Result this week.
||
*
||
*
|-

|-
|Pengqi Li
||
* Urgent tasks with Qiang Brother
** Handbook checking and revisions, PPT development
* Modified the paper for "Design course"
* Courses and report(Tuesday, Saturday, Today)
||
*
||
*
|-

|-
|Wan Lin
||
* Ablation experiments [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
* Experiments for diff. Transformer layers
||
*
||
*
|-

|-
|Tianhao Wang
||
* Huawei project things
** test report
** vggsound 44 classes model training
* Huawei interview
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* Sound separation
** familiar with the code of our model
* check and revise the AI handbooks
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* finished diarization test in vox-e (result is normal)
* read paper
||
*
||
*
|-

|-
|Jiaying Wang
||
* use Whisper replace the CTC calculation module: reading paper and preparing code.
* modify the bp strategy of the current CTC loss.
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi Agent System
** debug about new long-term short-term trade policy
* Huawei AED
** new dataset training (77% from Audio Set, 33% from Huawei) and some bad case study
||
*
||
*
|-

|-
|Wenqiang Du
||
* Summit middl handbook
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis review
||
*
||
*
|-
|Yue Gu
||
* almost finish 5 speakers'exps，plan to write the paper
* plan to give the presentation after accomplishing the introduction
* proofread the AI handbooks
||
*
||
*
|-
|-
|Qi Qu
||
* NPU (MR536) models benchmark regarding speed [https://b30lttjm7l.feishu.cn/docx/IygUdJ16kofcg2xEPnEc7lNsnqb?from=from_copylink].
* KWS models for Chaoshan dialect released.
||
*
||
*
|-

2025-03-24

2025-03-24T10:49:39Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* revisit cohort conditional chain overlap asr and conduct experiments (model is in training)
* thesis
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Complete some tests required by Huawei[https://z1et6d3xtb.feishu.cn/wiki/OXubwl2fIip91vkYsgMc1duhnLd]
||
*
||
*
|-

|-
|Junming Yuan
||
* Rewrite MT-HuBERT paper (1/5)
* AI practice handbook design of primary school and middle school(contents finished)
||
*
||
*
|-

|-
|Xiaolou Li
||
* VSR big data training (5500h) and debug
* Reproducing LipVoicer on Mandarin (overfitting now, still finding the best hyperparameter)
* Writing the VTS test document
||
*
||
*
|-

|-
|Zehua Liu
||
*GA VTS test document writing
*VTS paper Reading and Sharing this Friday
||
*
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University middle School.
* write paper for "Design course"
||
*
||
*
|-

|-
|Wan Lin
||
* run ablation experiments [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
||
*
||
*
|-

|-
|Tianhao Wang
||
* dragon05 data transfer
* 4-mix & 5-mix training
* huawei project things
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* Sound separation
** baseline: there are some bugs in previous code that resulting in low test results, so I modified the code and retrained it
* filter testing data for Huawei project
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* improve code efficiency, continue to test diarization on vox-e
* read paper
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Summit AI primary handbook
* AI primary handbook's PPT (44/44),Need to Check
* Check AI middl handbook(108/278)
||
*
||
*
|-

|-
|Yang Wei
||
* Text enroll kws model adaptation with keyword label from decoding and non-linear adaptation layer. (in progress)
||
*
||
*
|-

|-
|Turi
||
* Completed writing thesis (Will revise this week)
||
*
||
*
|-
|Yue Gu
||
* FIP-based personality-gated adaptation with synthetic data for personal ASR, almost finish preview evaluation exps.
* read 2 TTS-augment papers and reivew an interspeech2025 paper
||
*
||
*
|-
|-
|Qi Qu
||
* Text-enroll KWS: different window shifts (e.g. one at 100ms and the other at 200ms) for different keywords.
* VPR experiment: finding suitable thresholds for defined use cases.
||
*
||
*
|-

2025-03-17

2025-03-17T10:55:40Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook refinment for the college version.

||
*
||
*
|-

|-
|Lantian Li
||
* Submit of the AI-Graph EN version
* Review Master theses
||
*
||
*
|-

|-
|Ying Shi
||
* thesis
* some school-related stuff
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Complete the training and experimentation of the TSE-IRA model[https://z1et6d3xtb.feishu.cn/wiki/OXubwl2fIip91vkYsgMc1duhnLd] for Huawei, and will write the report
||
*
||
*
|-

|-
|Junming Yuan
||
* prepare report
* double-check middle-school AI handbook(1/7)
||
* write paper
* double-check middle-school AI handbook
* AI practice handbook design of primary school and middle school
||
*
|-

|-
|Xiaolou Li
||
* Writing NSFC document
* Find solutions for the low training speed with hybrid data disk
||
*
||
*
|-

|-
|Zehua Liu
||
*Writing NFSC document
*Reading VTS paper
||
*
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University middle School.
* Add references to the middle handbook(Done)
* Check middle handbook(1/3)
||
*
||
*
|-

|-
|Wan Lin
||
* train ns(multi-enroll) for voxblink1+voxceleb2: perform better than before, but still worse than vc2-only
* retrain multi-scenario ns(multi-enroll) model: consistently better than before now [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
* run ablation experiments: w/o multi-enroll/multi-head/two encoder
||
*
||
*
|-

|-
|Tianhao Wang
||
* sound sep subset training & testing
* huawei interview
* professor Guo's project investigation
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* Sound separation
** train AudioSep on subset data and evaluate it with our testing dataset
** read a paper about USS, try to test this model using our test set firstly(adjusting code in progress)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Finished diarization test in vox-o
** continue in vox-e
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook V3.1
* Check middle handbook
* We have Dragon05 (A6000 * 6)
||
*
||
*
|-

|-
|Yang Wei
||
* Text enroll kws adaptation experiment on cross lingual data. (not work. trying it with more complex adaptation layer)
||
*
||
*
|-

|-
|Turi
||
* Prepared ppt and poster for ICASSP2025
* Did presentation video and submitted with poster
||
*
||
*
|-
|Yue Gu
||
* synthesize 100h for each target spk and use KL loss as the regular term, the CER of target speakers reduce 10%.
||
*
||
*
|-
|-
|Qi Qu
||
* Android demo of Text-enroll KWS model.
||
*
||
*
|-

2025-03-17

2025-03-17T10:33:54Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook refinment for the college version.

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* thesis
* some school-related stuff
||
*
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* prepare report
* double-check middle-school AI handbook(1/7)
||
* write paper
* double-check middle-school AI handbook
* AI practice handbook design of primary school and middle school
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Finished diarization test in vox-o
** continue in vox-e
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook V3.1
* Check middle handbook
* We have Dragon05 (A6000 * 6)
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Prepared ppt and poster for ICASSP2025
* Presentation video
||
*
||
*
|-
|Yue Gu
||
* synthesize 100h for each target spk and use KL loss as the regular term, the CER of target speakers reduce 10%.
||
*
||
*
|-
|-
|Qi Qu
||
* Android demo of Text-enroll KWS model.
||
*
||
*
|-

2025-03-10

2025-03-10T10:57:22Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||

* Revise AI textbook of the colleage version

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Compare Ascend and Nvidia
** Performance: Clean ASR task 20epochs WER 6.91% : 7.02% (Ascend vs Nvidia)
** Speed: Nvidia is one time faster than Ascend
* Start think about my thesis
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* Pretraining work:
** MT-HuBERT & Cocktail-HuBERT will be finished next week.
** Get a set of comparable finetuning results(15/5/3-shot) for each pretrain model at the 400K training step.[https://z1et6d3xtb.feishu.cn/docx/ElAKdh07GoD8qKxGFLfc3seAnOh]
* Check and add reference for AI junior high school handbook(1/2).(Done)
||
*
||
*
|-

|-
|Xiaolou Li
||
* Writing NFSC document
* VSR training (1500 h) already have some result
** cnvsrc-single valid 300: 29.47%
** cnvsrc-multi valid: 31.60%
** webVideo valid: 15.54%
* Finished producing pseudo-label for CVS3(4000h)
||
*
||
*
|-

|-
|Zehua Liu
||
*Writing NFSC document
*Lora finetune VLM(both Encoder and LLM Decoder) result seem not very well(maybe need parameeter adjustment)
*Pretrained VSR Encoder + VLM(Decoder) seems better than Normal LM
||
*Design VTS architecture and implement it
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University Junior High School.
* Add references to the handbook(junior high school version 1/2)(Done).
||
*
||
*
|-

|-
|Wan Lin
||
* Supply NS experiments [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
* Help xiaochen reproduce the diarization SV method
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* speaker diarization baseline for NS (mix test: baseline EER 15.972% -> 12.983%) others still testing...
* make ppt about scaling law on speaker volume.
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi Agent Investment
** use Top 31 stocks in 11 sector to do portfolio for better correlation with input news (no excess return)
** analysis the trading decision
* Huawei AED
** smallest model to keep AUC excess 0.9
** split inference into two phase (Phase 1: Human Voice vs None Human Voice, Phase 2: Speech vs Other Human Voice) with two smaller model
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook V3.0（Done）
** Add reference(80%)
||
*
||
*
|-

|-
|Yang Wei
||
* Adapt text enroll kws model with synthesized dialect data.(recall: 83% -> 94%)[https://z1et6d3xtb.feishu.cn/docx/WFBJdF3D0o6w6bxHCJBcn9DIndg]
||
*
||
*
|-

|-
|Turi
||
* Finetuned Llama3 on Oromo text (pretrain)
* Experiment to use it as LM for ASR failed, 100%+ WER
||
*
||
*
|-
|Yue Gu
||
* a 0.4% CER reduction has achieved for one spk, but no improvement was discovered on other spks. I'm still do some exps.
* restart the synthetic-data related exps, try to fill the gap between synthetic data and real data on the output distribution of model.
||
*
||
*
|-
|-
|Qi Qu
||
* Technical investigation on Visual Event Detection.
* Experiment on annotating and auditing audio with Audio LLM: insufficient VRAM; poor I/O in CPU/GPU hybrid mode.
||
*
||
*
|-

2025-03-10

2025-03-10T10:56:38Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||

* Revise AI textbook of the colleage version

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Compare Ascend and Nvidia
** Performance: Clean ASR task 20epochs WER 6.91% : 7.02% (Ascend vs Nvidia)
** Speed: Nvidia is one time faster than Ascend
* Start think about my thesis
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* Pretraining work:
** MT-HuBERT & Cocktail-HuBERT will be finished next week.
** Get a set of comparable finetuning results(15/5/3-shot) for each pretrain model at the 400K training step.[https://z1et6d3xtb.feishu.cn/docx/ElAKdh07GoD8qKxGFLfc3seAnOh]
* Check and add reference for AI junior high school handbook(1/2).(Done)
||
*
||
*
|-

|-
|Xiaolou Li
||
* Writing NFSC document
* VSR training (1500 h) already have some result
** cnvsrc-single valid 300: 29.47%
** cnvsrc-multi valid: 31.60%
** webVideo valid: 15.54%
* Finished producing pseudo-label for CVS3(4000h)
||
*
||
*
|-

|-
|Zehua Liu
||
*Writing NFSC document
*Lora finetune VLM(both Encoder and LLM Decoder) result seem not very well(maybe need parameeter adjustment)
*Pretrained VSR Encoder + VLM(Decoder) seems better than Normal LM
||
*Design VTS architecture and implement it
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University Junior High School.
* Add references to the handbook(junior high school version 1/2)(Done).
||
*
||
*
|-

|-
|Wan Lin
||
* Supply NS experiments [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
* Help xiaochen reproduce the diarization SV method
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* speaker diarization baseline for NS (mix test: baseline EER 15.972% -> 12.983%) others still testing...
* make ppt about scaling law on speaker volume.
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi Agent Investment
** use Top 31 stocks in 11 sector to do portfolio for better correlation with input news (no excess return)
** analysis the trading decision
* Huawei AED
** smallest model to keep AUC excess 0.9
** split inference into two phase (Phase 1: Human Voice vs None Human Voice, Phase 2: Speech vs Other Human Voice) with two smaller model
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook V3.0（Done）
** Add reference(80%)
||
*
||
*
|-

|-
|Yang Wei
||
* Adapt text enroll kws model with synthesized dialect data.(recall: 83% -> 94%)[https://z1et6d3xtb.feishu.cn/docx/WFBJdF3D0o6w6bxHCJBcn9DIndg]
||
*
||
*
|-

|-
|Turi
||
* Finetuned Llama3 on Oromo text (pretrain)
* Experiment to use it as LM failed, 100%+ WER
||
*
||
*
|-
|Yue Gu
||
* a 0.4% CER reduction has achieved for one spk, but no improvement was discovered on other spks. I'm still do some exps.
* restart the synthetic-data related exps, try to fill the gap between synthetic data and real data on the output distribution of model.
||
*
||
*
|-
|-
|Qi Qu
||
* Technical investigation on Visual Event Detection.
* Experiment on annotating and auditing audio with Audio LLM: insufficient VRAM; poor I/O in CPU/GPU hybrid mode.
||
*
||
*
|-

2025-03-10

2025-03-10T10:54:47Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||

* Revise AI textbook of the colleage version

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Compare Ascend and Nvidia
** Performance: Clean ASR task 20epochs WER 6.91% : 7.02% (Ascend vs Nvidia)
** Speed: Nvidia is one time faster than Ascend
* Start think about my thesis
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* Pretraining work:
** MT-HuBERT & Cocktail-HuBERT will be finished next week.
** Get a set of comparable finetuning results(15/5/3-shot) for each pretrain model at the 400K training step.[https://z1et6d3xtb.feishu.cn/docx/ElAKdh07GoD8qKxGFLfc3seAnOh]
* Check and add reference for AI junior high school handbook(1/2).(Done)
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Writing NFSC document
*Lora finetune VLM(both Encoder and LLM Decoder) result seem not very well(maybe need parameeter adjustment)
*Pretrained VSR Encoder + VLM(Decoder) seems better than Normal LM
||
*Design VTS architecture and implement it
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University Junior High School.
* Add references to the handbook(junior high school version 1/2)(Done).
||
*
||
*
|-

|-
|Wan Lin
||
* Supply NS experiments [https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh?from=from_copylink]
* Help xiaochen reproduce the diarization SV method
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* speaker diarization baseline for NS (mix test: baseline EER 15.972% -> 12.983%) others still testing...
* make ppt about scaling law on speaker volume.
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi Agent Investment
** use Top 31 stocks in 11 sector to do portfolio for better correlation with input news (no excess return)
** analysis the trading decision
* Huawei AED
** smallest model to keep AUC excess 0.9
** split inference into two phase (Phase 1: Human Voice vs None Human Voice, Phase 2: Speech vs Other Human Voice) with two smaller model
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook V3.0（Done）
** Add reference(80%)
||
*
||
*
|-

|-
|Yang Wei
||
* Adapt text enroll kws model with synthesized dialect data.(recall: 83% -> 94%)[https://z1et6d3xtb.feishu.cn/docx/WFBJdF3D0o6w6bxHCJBcn9DIndg]
||
*
||
*
|-

|-
|Turi
||
*
||
*
||
*
|-
|Yue Gu
||
* a 0.4% CER reduction has achieved for one spk, but no improvement was discovered on other spks. I'm still do some exps.
* restart the synthetic-data related exps, try to fill the gap between synthetic data and real data on the output distribution of model.
||
*
||
*
|-
|-
|Qi Qu
||
* Technical investigation on Visual Event Detection.
* Experiment on annotating and auditing audio with Audio LLM: insufficient VRAM; poor I/O in CPU/GPU hybrid mode.
||
*
||
*
|-

2025-03-03

2025-03-03T10:53:35Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Prepare Ascend Sever environment
* training Conditional Chain overlap ASR model with Hierachical-Transformer [https://z1et6d3xtb.feishu.cn/docx/UPxGdBl4Zo6JapxHmsXcI14Pngg?from=from_copylink here]
||
*
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* Finish MPC-HuBERT pretrain.
* Double check the related experimental code.
** MT-HuBERT(in progress) & Cocktail-HuBERT need re-pretrain.
** The results of other baseline in [https://z1et6d3xtb.feishu.cn/docx/VThUd30RPoTBR4xOiKYc4gQTnsb here]
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Paper Reading and Sharing in last Friday
*Writing Vision Language Model code
*Writing NSFC document
||
*
||
*
|-

|-
|Pengqi Li
||
* Prepare the AI course for Tsinghua University Junior High School.
* Using t-SNE to visualize the factorized content vector.
** Next step is to color(speaker information importance or not) each point.
||
*
||
*
|-

|-
|Wan Lin
||
* try some methods for clean performance（no improvement）
* supply experiments for other tests
||
*
||
*
|-

|-
|Tianhao Wang
||
* sound separation: 2-mix and 3-mix model training
* weekly report
||
* subset data training
||
*
|-

|-
|Xiaoxue Luo
||
* generation of multi-mix audio data and did some test experiments.
* read papers
||
*
||
*
|-

|-
|Zhenyu Zhou
||
* finish graduation thesis
||
*
||
*
|-

|-
|Junhui Chen
||
* Reproducing speaker diarization method for NS (debugging...)
* read paper
||
*
||
*
|-

|-
|Jiaying Wang
||
* debug ctc loss part[https://z1et6d3xtb.feishu.cn/docx/TUHldiaoQoYBqux7JEhcaCXenzh]
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Primary handbook's PPT (24/44)
* Continue to check Primary and middle handbook(Completed this week)
* Speech cloning sample for the company

||
*
||
*
|-

|-
|Yang Wei
||
* Tuning text enroll kws model for dialect data with linear layer. (recall: 65%->85%->94%)
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
* Result with LM[https://z1et6d3xtb.feishu.cn/docx/JvDsd8zR4oMwnyxQEQdckpMjn7m?from=from_copylink]
||
*
||
*
|-
|Yue Gu
||
* finish some exps, but nothing is improved.
* finish a proposal，I will present it recently
||
*
||
*
|-
|-
|Qi Qu
||
* Applying pre-prod eval routine on text-enroll KWS models: the ideal thresholds for each keyword vary significantly.
||
*
||
*
|-

2025-02-24

2025-02-24T10:39:03Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional-Chain ASR
** Design frame level condition fusion method(The model is in training.)
||
*
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Primary handbook's PPT (17+4/44)
** Related xuean(17/44)
* Investigate cross lingual voice cloning
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Added Language model to ASR
** Evaluation with LM shows improvement [https://z1et6d3xtb.feishu.cn/docx/JvDsd8zR4oMwnyxQEQdckpMjn7m?from=from_copylink]
* Thesis writing
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-02-24

2025-02-24T10:37:59Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional-Chain ASR
** Design frame level condition fusion method(The model is in training.)
||
*
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Primary handbook's PPT (17+4/44)
** Related xuean(17/44)
* Investigate cross lingual voice cloning
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Added Language model to ASR
** Evaluation with LM shows improvement [](https://z1et6d3xtb.feishu.cn/docx/JvDsd8zR4oMwnyxQEQdckpMjn7m?from=from_copylink)
* Thesis writing
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-02-17

2025-02-17T10:52:10Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Middle-School education PPT

||
*
||
*
|-

|-
|Lantian Li
||
* Final version proofreading of the high-school book (17/40)
* Polish IS2025 papers.
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional chain overlap asr
** pooling order
** padding chain
** augmentation padding
** [https://z1et6d3xtb.feishu.cn/docx/Wg3ldbCKeoKhBzxZ2lccTO8ZnTf?from=from_copylink here]
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Completed the revision of the second edition of the paper
||
*
||
*
|-

|-
|Junming Yuan
||
* The results of MT-Hubert/Cocktail-Hubert/Hubert on LS960[https://z1et6d3xtb.feishu.cn/docx/DRICd10VCodDfOxXbDMcI97snod]
**15-shot finetuning(clean/Mixup/MT):MT-HuBERT > Cocktail-Hubert > Hubert
** MPC-Hubert is still in the training queue
||
*
||
*
|-

|-
|Xiaolou Li
||
* Paper modification
* Data processing (2000h / 4000h)
* Some new demand in data collection server
* Paper reading
||
*
||
*
|-

|-
|Zehua Liu
||
*CNVSRC 2024 and VSR-LLM paper writing and revise, and do some relevant Experiments.
||
*
||
*
|-

|-
|Pengqi Li
||
* XAI of Speaker Verification[https://z1et6d3xtb.feishu.cn/docx/QnZvdrNAboVCCVxgVRzcpsgRn6d]
** some experiment looks like successful
** Writing paper same time(30%)
*Report recently work on Friday
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* Check the pictures in AI high school handbook,will be completed in these days
* The learning and production of AI daily signature
||
*
||
*
|-

|-
|Zhenyu Zhou
||
* Graduation article
* interspeech paper with zhenghai
* code double check with jiaying
||
*
||
*
|-

|-
|Junhui Chen
||
* Read paper about e2e SV / loss (little useful content for NS).
* Continue to think and try other tricks, but no meaningful results yet.
||
*
||
*
|-

|-
|Jiaying Wang
||
* conditional chain with ctc experiment
** loss decreased slowly in the previous few epochs, almost stop at around -3
** checked the model code: no bug
** test results of these epoch to determine the problem
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*Continue to check AI primary handbook（done）
*Continue to check AI middle handbook（done）
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
* Trained the language model, some bugs when decoding with LM
||
*
||
*
|-
|Yue Gu
||
* fine-grained personality-gating method：finish the code and model training，ready to test
||
*
||
*
|-
|-
|Qi Qu
||
* Successful porting of text-enroll KWS models (hybrid quantization) to mr536 w/ low precision loss, same RT as previous version (which shows drastic precision loss). [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe?from=from_copylink]
||
*
||
*
|-

2025-02-17

2025-02-17T10:51:43Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Middle-School education PPT

||
*
||
*
|-

|-
|Lantian Li
||
* Final version proofreading of the high-school book (17/40)
* Polish IS2025 papers.
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional chain overlap asr
** pooling order
** padding chain
** augmentation padding
** [https://z1et6d3xtb.feishu.cn/docx/Wg3ldbCKeoKhBzxZ2lccTO8ZnTf?from=from_copylink here]
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Completed the revision of the second edition of the paper
||
*
||
*
|-

|-
|Junming Yuan
||
* The results of MT-Hubert/Cocktail-Hubert/Hubert on LS960[https://z1et6d3xtb.feishu.cn/docx/DRICd10VCodDfOxXbDMcI97snod]
**15-shot finetuning(clean/Mixup/MT):MT-HuBERT > Cocktail-Hubert > Hubert
** MPC-Hubert is still in the training queue
||
*
||
*
|-

|-
|Xiaolou Li
||
* Paper modification
* Data processing (2000h / 4000h)
* Some new demand in data collection server
* Paper reading
||
*
||
*
|-

|-
|Zehua Liu
||
*CNVSRC 2024 and VSR-LLM paper writing and revise, and do some relevant Experiments.
||
*
||
*
|-

|-
|Pengqi Li
||
* XAI of Speaker Verification[https://z1et6d3xtb.feishu.cn/docx/QnZvdrNAboVCCVxgVRzcpsgRn6d]
** some experiment looks like successful
** Writing paper same time(30%)
*Report recently work on Friday
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* Check the pictures in AI high school handbook,will be completed in these days
* The learning and production of AI daily signature
||
*
||
*
|-

|-
|Zhenyu Zhou
||
* Graduation article
* interspeech paper with zhenghai
* code double check with jiaying
||
*
||
*
|-

|-
|Junhui Chen
||
* Read paper about e2e SV / loss (little useful content for NS).
* Continue to think and try other tricks, but no meaningful results yet.
||
*
||
*
|-

|-
|Jiaying Wang
||
* conditional chain with ctc experiment
** loss decreased slowly in the previous few epochs, almost stop at around -3
** checked the model code: no bug
** test results of these epoch to determine the problem
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*Continue to check AI primary handbook（done）
*Continue to check AI middle handbook（done）
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
* Trained the language model, some when decoding with LM
||
*
||
*
|-
|Yue Gu
||
* fine-grained personality-gating method：finish the code and model training，ready to test
||
*
||
*
|-
|-
|Qi Qu
||
* Successful porting of text-enroll KWS models (hybrid quantization) to mr536 w/ low precision loss, same RT as previous version (which shows drastic precision loss). [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe?from=from_copylink]
||
*
||
*
|-

2025-02-10

2025-02-10T10:42:52Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
*
||
*
|-

|-
|Zhenghai You
||
* Revise the Is2025 paper to the second edition
* Reading some papers on SE/SS/TSE refiner
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* High-school PPT and Jiaoan(5)
||
*
||
*
|-

|-
|Wan Lin
||
* Revised NS paper for IS2025 [https://z1et6d3xtb.feishu.cn/docx/PqtDdWPSwomSDexW2oUc2M4qnrc?from=from_copylink] (still lack of experimental results)
* Train multi-scenario model (need parameters adjusting, still in training)
* Try other revised-BCE loss(failed)
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* faster test code for NS (Vox-O 25min -> 5min)
* writing paper for is25
* (On plane to Beijing)
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* refine prompt
* add backtesting report to debate report
* trying stronger model (DeepSeek R1 70B)
||
* introduce more trading groups that are mutually opposed.
||
*
|-

|-
|Wenqiang Du
||
* AI primary handbook PPT （16/35）
* Check AI primary handbook
* Check AI middle handbook
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Text data collection
* Adding Language Model to ASR (not successful yet)
||
*
||
*
|-
|Yue Gu
||
* almost finish phone-level gaussian model training
* according to the badcase analysis，I‘m designing a new framework for fine-grained speaker adaptation
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2025-01-20

2025-01-20T10:43:36Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI Graph high-school version, still on going

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Huawei TSE online demo design with Zhenghai
* Continue work on conditional overlap ASR model
||
*
*
||
*
|-

|-
|Zhenghai You
||
* TSE online service for Huawei project demonstration[https://huggingface.co/spaces/swc2/Target-speaker-extraction?logs=container]
** There are some channel issues, Poor performance when using some laptops or new launch Huawei phones
** Guess it may be caused by some hardware denoise
* BUPT host relocation
** Still offline
* Retraining for the best results in the paper experiment (2/10),and plan to release a version for IS2025 in the next two days at first
||
*
||
*
|-

|-
|Junming Yuan
||
* Go on MT-HuBERT paper writing(4/5)
* Go on model pretraining
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Interspeech Paper Writing
*Data checking
*Code writing for loading LLM
||
*
||
*
|-

|-
|Pengqi Li
||
* IS 25(Fail temporarily and keep a record.)
* Continue verifying the previous hypothesis
||
*
||
*
|-

|-
|Wan Lin
||
* paper reading
* NS experiments, but no new improvement now
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* Find a loss that may be helpful for NS[https://z1et6d3xtb.feishu.cn/docx/MxBNdPbLao0tsoxkBVCcUgUoneh#share-IpeLdu4PIodd3JxvjkccasOunJR], the code is ready.
||
*
||
*
|-

|-
|Jiaying Wang
||
* IS paper introduction part（checking with zhenyu）
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi Agent Investment Result [https://z1et6d3xtb.feishu.cn/docx/MXDydzMuVoWHwXxAuxic0q3AnAd]
||
*
||
*
|-

|-
|Wenqiang Du
||
* AI primary handbook PPT （4/35）
||
*
||
*
|-

|-
|Yang Wei
||
* Project
** With the help of Qi Qu, finished zhongchuan ASR REST service APP.
* Text Enroll KWS
** Training on 500 hours data (use word level keyword sampling, add data from sinvxx)
||
*
||
*
|-

|-
|Turi
||
* Text collection for adding language model to Oromo ASR
||
*
||
*
|-
|Yue Gu
||
* Finegrained personality extractor is still in debug
* Check the middle school handbook
* Have been sick yesterday and today
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** Pre-prod routine work for zh48+qingdao20 models.
** Experiment on optimizing inference efficiency on Android devices failed: xnnpack acceleration does not work for dynamic shapes; quantization does not work either.
* Misc:
** C++/C libraries for KWS/AED processing released for mr536.
||
*
||
*
|-

2025-01-13

2025-01-13T10:57:09Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version, recheck
* Publication stuff

||
*
||
*
|-

|-
|Lantian Li
||
* Go on AI-Graph EN (49/50)
||
*
||
*
|-

|-
|Ying Shi
||
* Try to reproduce the CTC conditional-chain ASR baseline (failed)
* check code with PIT loss (code looks fine)
* Try to train a model that only recognizes the most dominant components (failed)
||
*
*
||
*
|-

|-
|Zhenghai You
||
* The first Chinese version of the paper [https://www.overleaf.com/read/hgttnggctwps#2facce]
* Online Demo UI Design for Huawei Project[https://huggingface.co/spaces/swc2/Target-speaker-extraction]
||
*
||
*
|-

|-
|Junming Yuan
||
* Check the high school AI handbook(done)
* Organize our photos
* The results of MT-Hubert on LS960[https://z1et6d3xtb.feishu.cn/docx/HUIad3uqgozpEyxIwuTcU9uInsb]
**2-mixed Test: 400K steps top-2 ACC: 74.63% ---> 1600K steps top-2 ACC: 79.43%
||
*
||
*
|-

|-
|Xiaolou Li
||
* Process server code update
* Data audit (until 15th Jan)
||
*
||
*
|-

|-
|Zehua Liu
||
*Paper Reading
*Writing Code for loading larger LLM（32B and 72B）
*Interspeech paper writing
*Collected Data Checking （With Xiaolou）
||
*
||
*
|-

|-
|Pengqi Li
||
* IS25 Proposal almost done
** writing the experimental part
** checks and analysis the result
* Go to the hospital for a follow-up and return lab.
||
*
||
*
|-

|-
|Wan Lin
||
* NS: Adopt multi-enroll, margin-bce and long-duration test to resnet+transformer model: EER 1.23%->1.13%
||
*
||
*
|-

|-
|Tianhao Wang
||
* writing the chain-based sound sep code
* test some SED pretraining under mix scenario
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* got a bad flu
* read papers
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
* IS2025 proposal[https://z1et6d3xtb.feishu.cn/docx/T8IvdtN32o8kHWxqcPvc5LsWnce]
||
*
||
*
|-

|-
|Yu Zhang
||
* Finish Multi Agent Investment pipeline debug, experiment still running (can get a draft result this week)
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook
** Related PPT and Jiaoan
*Some project cooperation
||
*
||
*
|-

|-
|Yang Wei
||
* Train Text enroll kws model (pretrain w ctc loss + finetune w/o ctc loss). Not success yet.
* Develop ASR REST service based on FunASR model.
||
*
||
*
|-

|-
|Turi
||
* Did experiment on Conformer CTC with nbpe=500 (26 to 18 WER)
* Refined ICASSP paper for final submission and submitted it
* Prepare for interview
||
*
||
*
|-
|Yue Gu
||
* design the fine-grained personality extractor to produce the phone-level voice charactor similarity (code is in progress)
* check the primary school handbook
||
*
||
*
|-
|-
|Qi Qu
||
* AED:
** CED classifiers implemented on mr536 NPU.
* KWS:
** Training data collected and processed for Qingdao dialect, 20 keywords.
** Analysis of some prod FAs.
* Android demo and supporting backend services: KWS + ASR/MT -> instruction submission.
||
*
||
*
|-

2025-01-13

2025-01-13T10:56:32Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version, recheck
* Publication stuff

||
*
||
*
|-

|-
|Lantian Li
||
* Go on AI-Graph EN (49/50)
||
*
||
*
|-

|-
|Ying Shi
||
* Try to reproduce the CTC conditional-chain ASR baseline (failed)
* check code with PIT loss (code looks fine)
* Try to train a model that only recognizes the most dominant components (failed)
||
*
*
||
*
|-

|-
|Zhenghai You
||
* The first Chinese version of the paper [https://www.overleaf.com/read/hgttnggctwps#2facce]
* Online Demo UI Design for Huawei Project[https://huggingface.co/spaces/swc2/Target-speaker-extraction]
||
*
||
*
|-

|-
|Junming Yuan
||
* Check the high school AI handbook(done)
* Organize our photos
* The results of MT-Hubert on LS960[https://z1et6d3xtb.feishu.cn/docx/HUIad3uqgozpEyxIwuTcU9uInsb]
**2-mixed Test: 400K steps top-2 ACC: 74.63% ---> 1600K steps top-2 ACC: 79.43%
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Paper Reading
*Writing Code for loading larger LLM（32B and 72B）
*Interspeech paper writing
*Collected Data Checking （With Xiaolou）
||
*
||
*
|-

|-
|Pengqi Li
||
* IS25 Proposal almost done
** writing the experimental part
** checks and analysis the result
* Go to the hospital for a follow-up and return lab.
||
*
||
*
|-

|-
|Wan Lin
||
* NS: Adopt multi-enroll, margin-bce and long-duration test to resnet+transformer model: EER 1.23%->1.13%
||
*
||
*
|-

|-
|Tianhao Wang
||
* writing the chain-based sound sep code
* test some SED pretraining under mix scenario
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* got a bad flu
* read papers
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
* IS2025 proposal[https://z1et6d3xtb.feishu.cn/docx/T8IvdtN32o8kHWxqcPvc5LsWnce]
||
*
||
*
|-

|-
|Yu Zhang
||
* Finish Multi Agent Investment pipeline debug, experiment still running (can get a draft result this week)
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook
** Related PPT and Jiaoan
*Some project cooperation
||
*
||
*
|-

|-
|Yang Wei
||
* Train Text enroll kws model (pretrain w ctc loss + finetune w/o ctc loss). Not success yet.
* Develop ASR REST service based on FunASR model.
||
*
||
*
|-

|-
|Turi
||
* Did experiment on Conformer CTC with nbpe=500 (26 to 18 WER)
* Refined ICASSP paper for final submission and submitted it
* Interview
||
*
||
*
|-
|Yue Gu
||
* design the fine-grained personality extractor to produce the phone-level voice charactor similarity (code is in progress)
* check the primary school handbook
||
*
||
*
|-
|-
|Qi Qu
||
* AED:
** CED classifiers implemented on mr536 NPU.
* KWS:
** Training data collected and processed for Qingdao dialect, 20 keywords.
** Analysis of some prod FAs.
* Android demo and supporting backend services: KWS + ASR/MT -> instruction submission.
||
*
||
*
|-

2025-01-13

2025-01-13T10:47:48Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version, recheck
* Publication stuff

||
*
||
*
|-

|-
|Lantian Li
||
* Go on AI-Graph EN (49/50)
||
*
||
*
|-

|-
|Ying Shi
||
* Try to reproduce the CTC conditional-chain ASR baseline (failed)
* check code with PIT loss (code looks fine)
* Try to train a model that only recognizes the most dominant components (failed)
||
*
*
||
*
|-

|-
|Zhenghai You
||
* The first Chinese version of the paper [https://www.overleaf.com/read/hgttnggctwps#2facce]
* Online Demo UI Design for Huawei Project[https://huggingface.co/spaces/swc2/Target-speaker-extraction]
||
*
||
*
|-

|-
|Junming Yuan
||
* Check the high school AI handbook(done)
* Organize our photos
* The results of MT-Hubert on LS960[https://z1et6d3xtb.feishu.cn/docx/HUIad3uqgozpEyxIwuTcU9uInsb]
**2-mixed Test: 400K steps top-2 ACC: 74.63% ---> 1600K steps top-2 ACC: 79.43%
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* IS25 Proposal almost done
** writing the experimental part
** checks and analysis the result
* Go to the hospital for a follow-up and return lab.
||
*
||
*
|-

|-
|Wan Lin
||
* NS: Adopt multi-enroll, margin-bce and long-duration test to resnet+transformer model: EER 1.23%->1.13%
||
*
||
*
|-

|-
|Tianhao Wang
||
* writing the chain-based sound sep code
* test some SED pretraining under mix scenario
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* got a bad flu
* read papers
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
* IS2025 proposal[https://z1et6d3xtb.feishu.cn/docx/T8IvdtN32o8kHWxqcPvc5LsWnce]
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Check Primary handbook
** Related PPT and Jiaoan
*Some project cooperation
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Did experiment on Conformer CTC with nbpe=500 (26 to 18 WER)
* Refined ICASSP paper for final submission
* Interview
||
*
||
*
|-
|Yue Gu
||
* design the fine-grained personality extractor to produce the phone-level voice charactor similarity (code is in progress)
* check the primary school handbook
||
*
||
*
|-
|-
|Qi Qu
||
* AED:
** CED classifiers implemented on mr536 NPU.
* KWS:
** Training data collected and processed for Qingdao dialect, 20 keywords.
** Analysis of some prod FAs.
* Android demo and supporting backend services: KWS + ASR/MT -> instruction submission.
||
*
||
*
|-

2024-01-06

2025-01-06T10:53:58Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*

||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Analyze and summarize the previous experiments
* Project things
||
* continue on overlap asr task
* submit HUADANIAN challenge proposal
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Organize my previous Experiment Result[https://z1et6d3xtb.feishu.cn/docx/JBsidACDVojhCaxFQLbcCVbsnAc?from=from_copylink]
*Add correct loss, but code encounters some bugs, so i fix it and train again(still training)
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
* NS: multi-enroll+margin bce: 1.638% -> 1.521% (5w spk)
||
*
||
*
|-

|-
|Tianhao Wang
||
* Try using attention instead of FiLM:
** valid loss: query emb cross-attn: -8.065 vs. query feature cross-attn: -9.127 vs. FiLM: -10.054
** read a paper show that FiLM is better than attention in sound sep task
* Huawei project testing
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* final exam
* wrote the remaining course papers
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* finished all course homework and final exams
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Turi
||
* Midterm defense
* Assessment exam for application
* Experiment on improving Oromo ASR[https://z1et6d3xtb.feishu.cn/docx/JvDsd8zR4oMwnyxQEQdckpMjn7m?from=from_copylink]
||
*
||
*
|-
|Yue Gu
||
* report my recent work
* read paper about functionally invariant path, which was reported by Prof. Wang.
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** pre-prod routine work for zh48+yichang5 models.
** android demo backed with gRPC services deployed on peacock01.
* QAT: exploration with little progress.
||
*
||
*
|-

2024-12-30

2024-12-30T10:54:06Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
* Complete 2025 AI calendar
||
*
||
*
|-

|-
|Ying Shi
||
* Cosine-Guided Order VS Dominance Guided Order
** Token error rate and CTC-loss Confusion Matrix [https://z1et6d3xtb.feishu.cn/docx/HFJkdFp5XoM4wkxKb3Ucevhxnfb?from=from_copylink]
||
* Design Learnable order
* Think about how to use Condition ...
||
*
|-

|-
|Zhenghai You
||
* Paper reading for weekly report
* Supplement the SPK-AUG experiment and plan to start writing the paper[https://z1et6d3xtb.feishu.cn/docx/K4SYdNM2QoBFqOxFp7Hc8Z3Mnpg]
* Attempt to increase SSL loss on existing E2E-TSE{The current results are not good}
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-Hubert paper writing(1/2)
* Check the high school AI handbook(237/362)
||
*
||
*
|-

|-
|Xiaolou Li
||
* Data process server code upgrade, data transfer
* Experiment on 1500h (in training...)
* Odds and ends...Paper reading, Coursework, Calendar delivery...
||
* Upgrade the AV-HuBERT training code and test it
||
*
|-

|-
|Zehua Liu
||
*Train LRS3 exp for AlignVSR
*Iter Inference 7-times get result (43.88%) better than No Iter-Inference (45.74%)
*Use Weaker Encoder to generate corrupted text seems slightly better than before(43.88% < 44.54%)[https://z1et6d3xtb.feishu.cn/docx/JBsidACDVojhCaxFQLbcCVbsnAc?from=from_copylink]
||
*
||
*
|-

|-
|Pengqi Li
||
* Conduct a preliminary experiment for the proposal(IS25-XAI) (A clear doc must be produced by the end of this week.)
* Check the high school AI handbook(1/3)
||
*
||
*
|-

|-
|Wan Lin
||
* NS: margin BCE loss & multi-enroll training
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* prepare for the final exam
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Multi policy pipeline building (Done Tech/Sentiment policy generation)
* Some copyright stuff in Royal Flush
||
*
||
*
|-

|-
|Wenqiang Du
||
* Nothing~
||
*
||
*
|-

|-
|Yang Wei
||
* Prepare an ASR Java jar file and API doc for zhongchuan
||
*
||
*
|-

|-
|Turi
||
* Prepared finetuning code for MMS
** MMS from meta outperforms whisper and they provide adapters for every language supported which include Oromo language.
** Server is busy now, plan to do experiment on MMS
* Preparing ppt for midterm defense
||
* Midterm defense
||
*
|-
|Yue Gu
||
* read several papers about synthetic data for personalized ASR and do some exps. plan to report them on this friday or next monday.
||
*
||
*
|-
|-
|Qi Qu
||
* Quantization for NPU: metrics updated [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe].
* CED + classifier used as VAD: speech detection during inactive hours in dormitories.
* QAT (Quantization-Aware Training) exploration.
||
*
||
*
|-

2024-12-30

2024-12-30T10:51:33Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
* Complete 2025 AI calendar
||
*
||
*
|-

|-
|Ying Shi
||
* Cosine-Guided Order VS Dominance Guided Order
** Token error rate and CTC-loss Confusion Matrix [https://z1et6d3xtb.feishu.cn/docx/HFJkdFp5XoM4wkxKb3Ucevhxnfb?from=from_copylink]
||
* Design Learnable order
* Think about how to use Condition ...
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-Hubert paper writing(1/2)
* Check the high school AI handbook(237/362)
||
*
||
*
|-

|-
|Xiaolou Li
||
* Data process server code upgrade, data transfer
* Experiment on 1500h (in training...)
* Odds and ends...Paper reading, Coursework, Calendar delivery...
||
* Upgrade the AV-HuBERT training code and test it
||
*
|-

|-
|Zehua Liu
||
*Iter Inference 7-times get result (43.88%) better than No Iter-Inference (45.74%)
*Use Weaker Encoder to generate corrupted text seems slightly better than before(43.88% < 44.54%)[https://z1et6d3xtb.feishu.cn/docx/JBsidACDVojhCaxFQLbcCVbsnAc?from=from_copylink]
||
*
||
*
|-

|-
|Pengqi Li
||
* Conduct a preliminary experiment for the proposal(IS25-XAI) (A clear doc must be produced by the end of this week.)
* Check the high school AI handbook(1/3)
||
*
||
*
|-

|-
|Wan Lin
||
* NS: margin BCE loss & multi-enroll training
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* prepare for the final exam
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Nothing~
||
*
||
*
|-

|-
|Yang Wei
||
* Prepare an ASR Java jar file and API doc for zhongchuan
||
*
||
*
|-

|-
|Turi
||
* Prepare finetuning code for MMS
** MMS from meta outperforms whisper and they provide adapters for every language supported which include Oromo language.
** Server is busy now, plan to do experiment on MMS
* Prepare ppt for midterm defense
||
* Midterm defense
||
*
|-
|Yue Gu
||
* read several papers about synthetic data for personalized ASR and do some exps. plan to report them on this friday or next monday.
||
*
||
*
|-
|-
|Qi Qu
||
* Quantization for NPU: metrics updated [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe].
* CED + classifier used as VAD: speech detection during inactive hours in dormitories.
* QAT (Quantization-Aware Training) exploration.
||
*
||
*
|-

2024-12-23

2024-12-23T10:51:12Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version done (v1.1).
* 3rd training for Tsinghua middle school
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional-chain overlap asr
** Condition augmentation & condition detection (is running now)
** remove Decoder and investigate whether speech encoder can find the order only depend on CTC Loss and self-attention
* More tests for next diting (hope can be started within this week)
** google phone test
** Separation model test (thx tianhao, zhenyu)
* Final round interview
||
*
||
*
|-

|-
|Zhenghai You
||
* Huawei TSE Delivery Package
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-Hubert paper writing(1/4)
* Check the primary/middle school AI handbook
||
* Go on MT-Hubert paper writing
* Check the high-school AI handbook
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Check the primary/middle school AI handbook
* Conduct a preliminary experiment for the proposal(IS25-XAI)(But not ideal).
||
* Continue thinking about how to conduct the proposal.
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* weekly reading
* Huawei project things
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* write some course papers
||
* prepare for the final exam
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* 4 course homework/paper
* review for 3 math exams
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Huawei AED Delivery Package
* Royal Flush server base development environment setup
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
* Personal stuff (Setting up a website for African students Association in THU..)
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
* KWS: finding ideal thresholds for zh10 model (2024-12-13) optimized for exhibition scene.
* Quantization of Text-Enroll models: huge precision loss.
* NPU metrics: [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe?from=from_copylink report] (to be updated)
||
*
||
*
|-

2024-12-23

2024-12-23T10:50:08Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version done (v1.1).
* 3rd training for Tsinghua middle school
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional-chain overlap asr
** Condition augmentation & condition detection (is running now)
** remove Decoder and investigate whether speech encoder can find the order only depend on CTC Loss and self-attention
* More tests for next diting (hope can be started within this week)
** google phone test
** Separation model test (thx tianhao, zhenyu)
* Final round interview
||
*
||
*
|-

|-
|Zhenghai You
||
* Huawei TSE Delivery Package
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-Hubert paper writing(1/4)
* Check the primary/middle school AI handbook
||
* Go on MT-Hubert paper writing
* Check the high-school AI handbook
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Check the primary/middle school AI handbook
* Conduct a preliminary experiment for the proposal(IS25-XAI)(But not ideal).
||
* Continue thinking about how to conduct the proposal.
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* weekly reading
* Huawei project things
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* write some course papers
||
* prepare for the final exam
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* 4 course homework/paper
* review for 3 math exams
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
* Setting up a website for African students Association in THU
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
* KWS: finding ideal thresholds for zh10 model (2024-12-13) optimized for exhibition scene.
* Quantization of Text-Enroll models: huge precision loss.
* NPU metrics: [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe?from=from_copylink report] (to be updated)
||
*
||
*
|-

2024-12-23

2024-12-23T10:48:03Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI handbook high-school version done (v1.1).
* 3rd training for Tsinghua middle school
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Conditional-chain overlap asr
** Condition augmentation & condition detection (is running now)
** remove Decoder and investigate whether speech encoder can find the order only depend on CTC Loss and self-attention
* More tests for next diting (hope can be started within this week)
** google phone test
** Separation model test (thx tianhao, zhenyu)
* Final round interview
||
*
||
*
|-

|-
|Zhenghai You
||
* Huawei TSE Delivery Package
||
*
||
*
|-

|-
|Junming Yuan
||
* MT-Hubert paper writing(1/4)
* Check the primary/middle school AI handbook
||
* Go on MT-Hubert paper writing
* Check the high-school AI handbook
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Check the primary/middle school AI handbook
* Conduct a preliminary experiment for the proposal(IS25-XAI)(But not ideal).
||
* Continue thinking about how to conduct the proposal.
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* weekly reading
* Huawei project things
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* write some course papers
||
* prepare for the final exam
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* 4 course homework/paper
* review for 3 math exams
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Thesis writing
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
* KWS: finding ideal thresholds for zh10 model (2024-12-13) optimized for exhibition scene.
* Quantization of Text-Enroll models: huge precision loss.
* NPU metrics: [https://b30lttjm7l.feishu.cn/docx/WORcdiE1io86Agxg9hOcKIO0nwe?from=from_copylink report] (to be updated)
||
*
||
*
|-

2024-12-16

2024-12-16T10:35:02Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* High-school AI handbook recheck
* Talk on Hebei Education conference
* Paper review for TASLP

||
*
||
*
|-

|-
|Lantian Li
||
* Prepare 2025 Calendar (4th round, thx Xue/Lou)
* Review BUPT Master Viva Report (Done)
* Apply BUPT Innovation Scholar (Failed... Fxxk)
* Submit NSFC Annual Report (Done)
* Prepare some materials for another rubbish project...
||
* Go on AI-Graph EN Chapter 4 (37/50)
* Two HUAWEI projects delivery
||
*
|-

|-
|Ying Shi
||
* Revise the HUAWEI Project Proposal
** more test about Google App and write test report
* Prepare for the Interview
||
* Back to conditional chain-overlap asr model
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* paper reading
* Check the primary school version(8/10)
||
* Make Proposal for IS25 Special Session of XAI
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Evaluated ASR model using CER (6.08%) & Thesis writing
* Application for University
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-12-09

2024-12-09T10:54:07Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* High-school AI handbook (1/4)
* Examination sheet for AI challenge in Hebei

||
*
||
*
|-

|-
|Lantian Li
||
* Check high school AI book (Done)
* Polish 2025 Calendar (2nd round)
* BUPT master Viva talk
||
* AI-Graph EN Chapter 4 (37/50)
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
* Developing viva presentation
||
*
||
*
|-

|-
|Junming Yuan
||
* Add a Mixture Predictive Coding(MPC)-hubert experiment[https://z1et6d3xtb.feishu.cn/docx/Elild36IDoiGpfxYsYUcfoJwnWf].
** 2-mixed test: clean-Hubert < MPC-Hubert < cocktail-Hubert < MT-Hubert
||
* writing paper
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
* Preparing for proposal pre and report
||
*
||
*
|-

|-
|Zehua Liu
||
*Start term report and start-term material writing
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* start-term stuff
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* production of 2025 Calendar(v0.1 done)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*Mid term report and mid-term material writing
||
*
||
*
|-

|-
|Junhui Chen
||
* got a bad fever
* prepare for final exam
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* prepare thesis start term report
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Started writing thesis
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
* Finding ideal thresholds for model `zh48+chongqing15:v2-20241202`, which outperforms the previous one.
* SDK set up for `v851se`. Quantized KWS model works, ~17ms per inference. Some tiny bugs need further investigation.
||
*
||
*
|-

2024-12-09

2024-12-09T10:53:36Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* High-school AI handbook (1/4)
* Examination sheet for AI challenge in Hebei

||
*
||
*
|-

|-
|Lantian Li
||
* Check high school AI book (Done)
* Polish 2025 Calendar (2nd round)
* BUPT master Viva talk
||
* AI-Graph EN Chapter 4 (37/50)
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
* Developing viva presentation
||
*
||
*
|-

|-
|Junming Yuan
||
* Add a Mixture Predictive Coding(MPC)-hubert experiment[https://z1et6d3xtb.feishu.cn/docx/Elild36IDoiGpfxYsYUcfoJwnWf].
** 2-mixed test: clean-Hubert < MPC-Hubert < cocktail-Hubert < MT-Hubert
||
* writing paper
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
* Preparing for proposal pre and report
||
*
||
*
|-

|-
|Zehua Liu
||
*Start term report and start-term material writing
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* start-term stuff
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* production of 2025 Calendar(v0.1 done)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*Mid term report and mid-term material writing
||
*
||
*
|-

|-
|Junhui Chen
||
* got a bad fever
* prepare for final exam
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* prepare thesis start term report
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Started writing thesis
*
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
* Finding ideal thresholds for model `zh48+chongqing15:v2-20241202`, which outperforms the previous one.
* SDK set up for `v851se`. Quantized KWS model works, ~17ms per inference. Some tiny bugs need further investigation.
||
*
||
*
|-

2024-12-02

2024-12-02T10:52:14Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Some personal stuff
* Find some numerical issues of current conditional model (will be fixed in this week)
||
*
||
*
|-

|-
|Zhenghai You
||
* Huawei TSE IRA Demo[https://z1et6d3xtb.feishu.cn/docx/OyOKdtjGJoQBDxxEalScvEKynLg]
||
*
||
*
|-

|-
|Junming Yuan
||
* Some varient experiments with MT-HuBERT[https://z1et6d3xtb.feishu.cn/docx/NErQdl2Hkoa6hSx8qFrc4LQInzc]
* get familiar with text-enroll experiment
||
*
||
*
|-

|-
|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* Some test for out-of-domain text query gap [https://z1et6d3xtb.feishu.cn/docx/KprndYrp3oe8KJxedj8cD8Yfnvd]
* thesis proposal
||
* thesis proposal
||
*
|-

|-
|Xiaoxue Luo
||
* production of 2025 Daily Sign(5/12)
||
* production of 2025 Daily Sign(12/12)
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* thesis proposal
* paper reading about Multi-Agent Collaboration
** Cooperation in Different Roles
** Competition in the Same Role
||
*
||
*
|-

|-
|Wenqiang Du
||
* Project cooperation communication
** Tencent
** dongxin- technical report about vpr
** zhongchuan - local asr
** Heeryuan - In bidding about RAG

||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Paper reading
||
*
||
*
|-
|Yue Gu
||
* ablation study about adaptation data, badcase analysis [https://z1et6d3xtb.feishu.cn/wiki/VPZfwx53ei2zkgkSvPtcCiDSnVh?from=from_copylink]
* read several papers about synthetic data for data argumentation
||
*
||
*
|-
|-
|Qi Qu
||
* Finding ideal thresholds for `zh48+henan22` and `zh48+chongqing15` KWS models. `zh48+chongqing15` performs poorly and needs re-arranging word list and retraining.
* NPU development/test environment setup: mr536 and v851se (new).
||
*
||
*
|-

2024-11-25

2024-11-25T10:32:43Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* 2nd round of check for AI handbook middle school.
* Deal with pictures in AI handbook (primary & middle).
* Start to check AI handbook high school.
* Check AI book for Tianjin medical school.
||
*
||
*
|-

|-
|Lantian Li
||
* Complete my CSTR Report
* Go on AI-Graph EN Chapter 4
* Polish 2025 Daily Sign
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Rebutall writing
*Iterative training
**Baseline(45.99%) < Iter-1(45.53%) < Iter-2(45.00%) < Iter-3(44.85%)
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Huawei AED
** data aug & human annotated dataset [https://z1et6d3xtb.feishu.cn/wiki/AO2CwQC4gioaq6k1SkkcARBAn2f]
* Finance
** Paper reading, reproduce local Llama version of StockAgent [https://github.com/MingyuJ666/Stockagent] (a LLM based market simulation framework)
||
*
||
*
|-

|-
|Wenqiang Du
||
*Training of New language Models(HeNan)[https://z1et6d3xtb.feishu.cn/docx/R7uIdGnwBo69bqxXki8cn50cnfh?from=from_copylink]
*Training of New language Models(ChongQing)[https://z1et6d3xtb.feishu.cn/docx/FioOdh8Uqo83oCxAJRzcGXcRnae?from=from_copylink]

||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Paper reading
* ICASSP 2025 rebuttal
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-11-11

2024-11-11T10:46:33Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Tianjian AI book (done)
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* re-check some details from Cocktail HuBERT paper and prepared the code.
* paper reading
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*Attemp to add silence loss during training（seems like useless）
*Conditional Chain 2-5 mix results（still some bugs，the acc of speaker number is poor）[https://z1et6d3xtb.feishu.cn/docx/D2UQdxMBvojkF9xCXGfcFBLGned]
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* SocioDojo
** Single stock (TSLA) investment (still running)
* Investigate some Text guided LLM centric time-series forecaster and reproduce some of them (Time-LLM LLM-Process, AutoTimes), and some toy experiment about how prompt prefix influence the forecast result
||
*
||
*
|-

|-
|Wenqiang Du
||
* Training of New Dialect Models(Cantonese)
* Prepare the PPT for the competition
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* kws data preparation and checking some implementations
* Paper Reading about kws
||
*
||
*
|-
|Yue Gu
||
* use CosyVoice model to synthesize the target speaker utterance, which is employed as the supplement for target speaker adaptation. The adaptation exp is running.
* icassp 2025 paper review
* paper writing
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-11-11

2024-11-11T10:45:52Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Tianjian AI book (done)
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* re-check some details from Cocktail HuBERT paper and prepared the code.
* paper reading
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*Attemp to add silence loss during training（seems like useless）
*Conditional Chain 2-5 mix results（still some bugs，the acc of speaker number is poor）[https://z1et6d3xtb.feishu.cn/docx/D2UQdxMBvojkF9xCXGfcFBLGned]
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* SocioDojo
** Single stock (TSLA) investment (still running)
* Investigate some Text guided LLM centric time-series forecaster and reproduce some of them (Time-LLM LLM-Process, AutoTimes), and some toy experiment about how prompt prefix influence the forecast result
||
*
||
*
|-

|-
|Wenqiang Du
||
* Training of New Dialect Models(Cantonese)
* Prepare the PPT for the competition
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* kws data preparation and checking some implementations
* Paper Reading about kws
*
||
*
||
*
|-
|Yue Gu
||
* use CosyVoice model to synthesize the target speaker utterance, which is employed as the supplement for target speaker adaptation. The adaptation exp is running.
* icassp 2025 paper review
* paper writing
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-11-11

2024-11-11T10:45:17Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Tianjian AI book (done)
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* re-check some details from Cocktail HuBERT paper and prepared the code.
* paper reading
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*Attemp to add silence loss during training（seems like useless）
*Conditional Chain 2-5 mix results（still some bugs，the acc of speaker number is poor）[https://z1et6d3xtb.feishu.cn/docx/D2UQdxMBvojkF9xCXGfcFBLGned]
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* SocioDojo
** Single stock (TSLA) investment (still running)
* Investigate some Text guided LLM centric time-series forecaster and reproduce some of them (Time-LLM LLM-Process, AutoTimes), and some toy experiment about how prompt prefix influence the forecast result
||
*
||
*
|-

|-
|Wenqiang Du
||
* Training of New Dialect Models(Cantonese)
* Prepare the PPT for the competition
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* KWS data preparation and checked some implementations
* Paper Reading about kws
*
||
*
||
*
|-
|Yue Gu
||
* use CosyVoice model to synthesize the target speaker utterance, which is employed as the supplement for target speaker adaptation. The adaptation exp is running.
* icassp 2025 paper review
* paper writing
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-11-04

2024-11-04T10:42:48Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* paper reading
* prepare to reproduce cocktail HuBERT (in progress)
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
* Debug the Chinese VTS (in training already)
* Write the report of VTS project (main work)
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
* prepare the report
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
* NS with frame-level detection loss
** use silero-vad
** Model is training, seems EER decrease faster.
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* SocioDojo
** with cash ratio risk aware, and change information sources, seems have a decent risk control over Nasdaq 100 index [https://z1et6d3xtb.feishu.cn/docx/E6JHdDItzoUe1Ix7Bkfcp435n5b?from=from_copylink]
* Some paper reading and report in RoyalFlush, get some idea (mainly about LLM for time series task)
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* LoRA finetuning (Result is not good)
* Data cleaning
||
*
||
*
|-
|Yue Gu
||
* read several paper about speech tokenizer. I want to design a encoder, which processes different size feature frame and construct several different codebooks, to extract personality from the varing speech speed. It is still in progress.
* paper writing
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** Yi (Liangshan, Sichuan) dataset prepared for training; dataset to be annotated for testing.
** Experiments on model quantization for NPU devices: i16 quantization arrives at a balance between accuracy and efficiency (~2ms per inference, compared to ~250ms for non-quantized); more calibration data needed for further confirmation.
** Full-featured demo (recording + feature extraction + model inference) for NPU devices in development.
||
*
||
*
|-

2024-10-28

2024-10-28T10:50:10Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AI primary book done
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* revise the code about cohort-overlap asr [the training is in progress]
** Support arbitrary source mixing training
** Use the real hypothesis by Token error rate
||
*
||
*
|-

|-
|Zhenghai You
||
* Introduce more hard samples to improve model performance[https://z1et6d3xtb.feishu.cn/docx/CURxdy3tEorxkrxtjjqcdMaYnJg]
** SPK-AUG with same length: There is an improvement, but the SI-SDR decreases when hard sample rate increases
** Design more hard samples
||
*
||
*
|-

|-
|Junming Yuan
||
* The result of time-mask MT-HuBERT [https://z1et6d3xtb.feishu.cn/docx/P62VdvJ9ooDxogx1Q09cM8Zmnd2]
** A sad news
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*Reading Papper about In-Context-Learning in ASR
*Training model with Adaptive Time Mask
*Try In-Context-Learning with only previous sentence[https://z1et6d3xtb.feishu.cn/docx/JBsidACDVojhCaxFQLbcCVbsnAc?from=from_copylink]
*VTS Project Report starts
||
*
||
*
|-

|-
|Pengqi Li
||
*Consistency of TAO and LayerCAM
** Change TAO from input to final conv layer and obtain more consistency.(Aishell:0.93 in any model)
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* AudioSep (CLAP) 5-mix exps:
** text-query: SDR=4.978, SI-SDR=1.972
** audio-query: SDR=6.907, SI-SDR=5.058
** This results with the loudness limitation
||
* AudioSep (CLAP) without loudness limitation
* Project things
||
*
|-

|-
|Xiaoxue Luo
||
*Comparative experiment between AudioSep and baseline system(CLIPSep)
*Prepare the report
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*reproduce 5-mix speech Separation results：
**pit：2-mix：16.04 ；5-mix：6.87
**conditional：5-mix：5.38（40 epoch）
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* SocioDojo (still worse than Nasdaq100 baseline)
** Change information sources, from the perspective of the report generated by LLM, more new information sources will be referenced.
** Prompt Actuator to consider current cash ratio before investing (with out this, the asset ratio goes up to 100%, which leads to high risks, still running)
* Read some papers about integrating time series into LLM
||
*
||
*
|-

|-
|Wenqiang Du
||
* Prepare data,code and environment for Pro.Mijiti
||
*
||
*
|-

|-
|Yang Wei
||
* Train text enroll KWS model with Aibabel training data. Not work.
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Whisper-largev3 finetuning
** Freezing 20 layers of encoder achieved 9.75 WER. Vanilla finetuning 8.02 WER
||
*
||
*
|-
|Yue Gu
||
* seek suggestions from other authors. Many suggestions are conflicting, so I'm try to figure out the reasons and fix these issues.
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** Text-enroll models exported to ONNX.
** C/JNI libs built based on ONNX models and ready for on-device test.
||
*
||
*
|-

2024-10-21

2024-10-21T10:47:08Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Primary School AI hand book (20-30)
||
*
||
*
|-

|-
|Lantian Li
||
* AI-Graph EN (25/50)
* Complete CSTR intro report （11.18)
||
*
||
*
|-

|-
|Ying Shi
||
* Cohort-Overlap ASR
** condition on real decode result
** Design stop criterion
* Cohort-Speech separation
** several configs for Dual-path model
* [https://z1et6d3xtb.feishu.cn/docx/X3C4dFRoKo3PATxvXWUc40LOnye?from=from_copylink group work ]
||
*
||
*
|-

|-
|Zhenghai You
||
* Weekly report
||
*
||
*
|-

|-
|Junming Yuan
||
* The result of feat-mask/time-mask MT-HuBERT [https://z1et6d3xtb.feishu.cn/docx/IXBzdx2OvoDbvLxBLmKcjGJZnvb]
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Complete the final report of the doctoral innovation project(School)
* Exploring the Consistency of TAO and LayerCAM Results on different models and datasets.
** Conclusion and hypothesis[https://z1et6d3xtb.feishu.cn/docx/Elh0d3t2RoEcUOxoIh2cBQG5nCc]
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
* adjust the code of AudioSep (CLAP) to support multi-mix and audio-query (in training)
* some project testing
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*AudioSep reproduction
**evaluate the performance of AudioSep
**comparative experiment between AudioSep and baseline system(CLIPSep)
***adjusting the code
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Participated in an AI competition
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Whisper finetuning on sagalee
** with encoder frozen, whisper-large-v3 (20.5 WER)
* Finetuning LLM
** Finetuned Qwen2.5-0.5B on conversation dataset translated from English to Oromo
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-10-21

2024-10-21T10:37:49Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Primary School AI hand book (20-30)
||
*
||
*
|-

|-
|Lantian Li
||
* AI-Graph EN (25/50)
* Complete CSTR intro report （11.18)
||
*
||
*
|-

|-
|Ying Shi
||
* Cohort-Overlap ASR
** condition on real decode result
** Design stop criterion
* Cohort-Speech separation
** several configs for Dual-path model
* [https://z1et6d3xtb.feishu.cn/docx/X3C4dFRoKo3PATxvXWUc40LOnye?from=from_copylink group work ]
||
*
||
*
|-

|-
|Zhenghai You
||
* Weekly report
||
*
||
*
|-

|-
|Junming Yuan
||
* The result of feat-mask/time-mask MT-HuBERT [https://z1et6d3xtb.feishu.cn/docx/IXBzdx2OvoDbvLxBLmKcjGJZnvb]
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Complete the final report of the doctoral innovation project(School)
* Exploring the Consistency of TAO and LayerCAM Results on different models and datasets.
** Conclusion and hypothesis[https://z1et6d3xtb.feishu.cn/docx/Elh0d3t2RoEcUOxoIh2cBQG5nCc]
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*AudioSep reproduction
**evaluate the performance of AudioSep
**comparison experiment between AudioSep and baseline system(CLIPSep)
***adjusting the code
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
* Participated in an AI competition
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Whisper finetuning on sagalee
** with encoder frozen, whisper-large-v3 (20.5 WER)
* Finetuning LLM
** Finetuned Qwen2.5-0.5B
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-10-14

2024-10-14T10:31:57Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|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)
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
* Evaluate TAO and LayerCAM(verification) reliability.
** Exploring the Consistency of TAO and LayerCAM Results on different models and datasets.
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*Paper reading about sound separation
*AudioSep reproduction
**Training time is too long -> replace with a small dataset(in training)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|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
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Whisper model finetuning[https://uestc.feishu.cn/docx/LsVmd6Mr3o90U2xVULOcImPmnjb?from=from_copylink]
||
*
||
*
|-
|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.
||
*
||
*
|-

Cslt-member-stu

2024-10-02T02:02:17Z

Turi：/* Turi Abu (途睿) */

==CSLT Students==

=== Chang Feng (冯畅) ===
[[文件:Fengchang.jpg | 200px]]
* 2019.9-
* Speaker Recognition

=== Renmiao Chen(陈仁苗) ===
[[文件:Chenrenmiao.jpg |200px]]
* PhD student
* 2021.09 -
* Speaker Recognition

=== Turi Abu (途睿) ===
[[文件:Resizedimg.jpg | 200px]]
* Master student
* 2023.9-
* Keyword Spotting, ASR
*[[媒体文件:Turi_biweekly_report.pptx|Bi-weekly report]]

==ASIP Students==

===Ying Shi (石颖）===
[[文件:Ying_shi.jpg|200px]]
* 2016.6.15 - 2017.7 (Intern)
* 2017.8 - 2020.1 (Engineer)
* 2020.1 -
* Speech processing
* [[媒体文件:Shiying_bi_weekly_report.ppt|Bi-weekly report]]
*[[媒体文件:DataSecurityAgreement_YingShi.jpg|DataSecurityAgreement]]

=== Pengqi Li (李鹏琦) ===
[[文件:Lipengqi.jpg |200px]]
* XJU
* 2020.9.30 -
* [[媒体文件:201228-Device_Generalization.pdf|Bi-weekly report]]
* Speaker recognition

=== Lily (毛丽旦) ===
[[文件:Maolidan.jpg]]
* XJU
* 2022.7.23 -
* Second language acquisition

=== Wan Lin (林琬) ===
[[文件:Linwan.jpg]]
* SZU
* 2022.7.23 -
* [[媒体文件:220722-AutoED_SRE_LW.pdf|Bi-weekly report]]
* Speaker recognition

=== Tianhao Wang (王天昊) ===
[[文件:WangTianhao.jpg]]
* HRBUST
* BUPT
* 2022.8.12 -
* [[媒体文件:Extremal_perturbations.pdf| Bi-weekly report]]
* Speaker recognition

=== Junhui Chen (陈俊卉) ===
[[文件:chenjunhui3.jpg]]
* BUPT
* 2022.11.04 -
* [[媒体文件:WeeklyReport.pdf| Bi-weekly report]]
* Speaker recognition

=== Zhenyu Zhou (周振宇) ===
[[文件: 周振宇.jpg]]
* BUPT
* 2022.11.04 -
* [[媒体文件:20221104_acoustic_training.pdf| Bi-weekly report]]
* Speaker recognition

=== Junming Yuan (袁浚铭) ===
[[文件:yuanjunming.jpg]]
* XJU
* 2023.8.24 -
* [[媒体文件:231204-Few-shot_KWS-yjm.pdf ‎| Bi-weekly report]]
* Speech processing

=== ZhengHai You (游正海) ===
[[文件:Youzhenghai_dealphoto.jpg]]
* BUPT
* 2023.4.22 -
* [[媒体文件:OC2023修改版3.0.pptx|Bi-weekly report]]
* Speech processing

Cslt-member-stu

2024-10-02T02:01:16Z

Turi：

==CSLT Students==

=== Chang Feng (冯畅) ===
[[文件:Fengchang.jpg | 200px]]
* 2019.9-
* Speaker Recognition

=== Renmiao Chen(陈仁苗) ===
[[文件:Chenrenmiao.jpg |200px]]
* PhD student
* 2021.09 -
* Speaker Recognition

=== Turi Abu (途睿) ===
[[文件:Resizedimg.jpg | 200px]]
* Master student
* 2023.9-
* Keyword Spotting, ASR
*[[媒体文件:Turi_biweekly_report.ppt|Bi-weekly report]]
==ASIP Students==

===Ying Shi (石颖）===
[[文件:Ying_shi.jpg|200px]]
* 2016.6.15 - 2017.7 (Intern)
* 2017.8 - 2020.1 (Engineer)
* 2020.1 -
* Speech processing
* [[媒体文件:Shiying_bi_weekly_report.ppt|Bi-weekly report]]
*[[媒体文件:DataSecurityAgreement_YingShi.jpg|DataSecurityAgreement]]

=== Pengqi Li (李鹏琦) ===
[[文件:Lipengqi.jpg |200px]]
* XJU
* 2020.9.30 -
* [[媒体文件:201228-Device_Generalization.pdf|Bi-weekly report]]
* Speaker recognition

=== Lily (毛丽旦) ===
[[文件:Maolidan.jpg]]
* XJU
* 2022.7.23 -
* Second language acquisition

=== Wan Lin (林琬) ===
[[文件:Linwan.jpg]]
* SZU
* 2022.7.23 -
* [[媒体文件:220722-AutoED_SRE_LW.pdf|Bi-weekly report]]
* Speaker recognition

=== Tianhao Wang (王天昊) ===
[[文件:WangTianhao.jpg]]
* HRBUST
* BUPT
* 2022.8.12 -
* [[媒体文件:Extremal_perturbations.pdf| Bi-weekly report]]
* Speaker recognition

=== Junhui Chen (陈俊卉) ===
[[文件:chenjunhui3.jpg]]
* BUPT
* 2022.11.04 -
* [[媒体文件:WeeklyReport.pdf| Bi-weekly report]]
* Speaker recognition

=== Zhenyu Zhou (周振宇) ===
[[文件: 周振宇.jpg]]
* BUPT
* 2022.11.04 -
* [[媒体文件:20221104_acoustic_training.pdf| Bi-weekly report]]
* Speaker recognition

=== Junming Yuan (袁浚铭) ===
[[文件:yuanjunming.jpg]]
* XJU
* 2023.8.24 -
* [[媒体文件:231204-Few-shot_KWS-yjm.pdf ‎| Bi-weekly report]]
* Speech processing

=== ZhengHai You (游正海) ===
[[文件:Youzhenghai_dealphoto.jpg]]
* BUPT
* 2023.4.22 -
* [[媒体文件:OC2023修改版3.0.pptx|Bi-weekly report]]
* Speech processing

文件:Turi biweekly report.pptx

2024-10-02T02:00:31Z

Turi：

2024-09-30

2024-09-30T10:50:43Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
* AI-Graph handbook v0.1
* AI-Graph EN (12/50)
* Huawei TiDing 3.0 - Model Quantization
* BUPT/AI-Radiance trivial things
||
*
||
*
|-

|-
|Ying Shi
||
* Add 4 kinds of negative sampling strategies Optimized Text-enroll KWS code
** (deletion, substitution, insertion, and shuffle) and verify them to ensure no bugs.
** Find that new negative sampling will increase the difficulty of training which indicates that only depending on positional embedding is not enough.
* Reproduce conditional chain overlap asr (Sequence to Multi-Sequence Learning via Conditional Chain Mapping for Mixture Signals)
** According to Jiaying's work the code released by the published paper can not work
** Write dominance-based conditional chain overlap asr by myself (in progress)
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
* Use MFA on LRS3 to cut it into small segments
* Use discrete embedding of avhubert in vsp-llm training (Still training)
* Some idea of align video feature and LLM (Dense Connector, CL methods)
* Handover the data collection and get familiar with the process
* Data Collection: 3138 h (need to re-check, DDL: 10.15)
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
* Voxblink1 model training and testing [https://z1et6d3xtb.feishu.cn/docx/BywjdkGvNou12sxQ4dAcxYa9noh?from=from_copylink]
||
*
||
*
|-

|-
|Tianhao Wang
||
* AudioSep reproduction
** problem: LAION CLAP needs 48kHz audio so the data needs to be up-resample
||
*
||
*
|-

|-
|Xiaoxue Luo
||
*AI-Graph High school handbook(v0.1)
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*Check primary school handbook(43/45)
*Release chinese and haining KWS model
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Segmented audios in dataset into individual words.
* Paper reading
||
*
||
*
|-
|Yue Gu
||
* Almost complete the revisions of my journal paper
||
*
||
*
|-
|-
|Qi Qu
||
* KWS
** Testing zh48 models on dataset of Mandarin Chinese w/ Guangdong accent: recall drops significantly.
* AED
** Evaluating third-party solution of baby crying detection.
* Misc.
** Preparing for live talk.
||
*
||
*
|-

2024-09-23

2024-09-23T10:48:31Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AIGraph high education version
* Prepare AIgraph Large Model version
* NMI paper publication staff
||
*
||
*
|-

|-
|Lantian Li
||
* AI-Graph EN (1/4)
* Huawei Project Proposal v1.0
* First Lesson on 24-fall AI Undergraduates
||
*
||
*
|-

|-
|Ying Shi
||
* Huawei project proposal
* Optimize the Text-enroll KWS code
** improve readability.
** remove redundant code.
||
*
||
*
|-

|-
|Zhenghai You
||
* Exploring the generality of spk aug on different data and structures[https://z1et6d3xtb.feishu.cn/docx/XXjPdjTjho7qwNxzEr7cTVjjnse]
||
*
||
*
|-

|-
|Junming Yuan
||
* double check mixed Hubert code：
** fix some bugs (time-mask.etc)
** time-mask vs. feat mask: (27.98%, 23.17%) vs.(23.19%, 25.99%)
** softmax+CE --> sigmoid+BCE still have problem.
||
*
||
*
|-

|-
|Xiaolou Li
||
* Writing VTS documents
* Paper Reading & Preparing for Report
* Exp on LRS3
** LLM: LLaMA2 -> LLaMA3.1 (30h ↓0.4%)
** Grouping LLaMA2: (443h ↑0.5%, 30h ↓2.5%)
* Rethinking the method to inject information (ablation study first)
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
* VC2 pre-train; VB1+VC2 mix-turning
** Data filter in VB1: 1.25% EER in vox1-o
* VB1 pre-train; VC2 fine-tuning
** VB1 pre-train: 2.61% EER in vox1-o
** VC2 fine-tuning: maybe couldn't reach better performance
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Dataset collection from THS
* Retraining R^2 SAC paper, with same env still failed (TCN ACC: 0.708, RECALL: 0.183), will check with Han this week
* Paper reading and some plan (report this Fri)
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
* Train text enroll KWS model with aishell and kespeech data.
* Prepare live broadcast
||
*
||
*
|-

|-
|Lily
||
* Prepare for holiday course（October 2nd、3rd） and online-course
* AI radiance's daily work
||
*
||
*
|-

|-
|Turi
||
* Trained conformer on Sagalee data excluding utterances containing digits
** Achieved 21.28% WER, 2.65 WER reduction
* Preparing KWS data from Sagalee dataset using MFA
||
*
||
*
|-
|Yue Gu
||
* paper writing
* open the code
* prepare for the presentation
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** Finding ideal thresholds for b0-models in predefined scenes: Chinese Mandarin, Cantonese, Uyghur and Kazakh.
** Finding ideal thresholds for b6-models with fixed b0-model thresholds.
* AED:
** Fixing parameters of Fbank feature extraction for CED and retraining classifiers.
||
*
||
*
|-

2024-09-23

2024-09-23T10:41:43Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* AIGraph high education version
* Prepare AIgraph Large Model version
* NMI paper publication staff
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
* Huawei project proposal
* Optimize the Text-enroll KWS code
** improve readability.
** remove redundant code.
||
*
||
*
|-

|-
|Zhenghai You
||
* Exploring the generality of spk aug on different data and structures[https://z1et6d3xtb.feishu.cn/docx/XXjPdjTjho7qwNxzEr7cTVjjnse]
||
*
||
*
|-

|-
|Junming Yuan
||
* double check mixed Hubert code：
** fix some bugs (time-mask.etc)
** time-mask vs. feat mask: (27.98%, 23.17%) vs.(23.19%, 25.99%)
** softmax+CE --> sigmoid+BCE still have problem.
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* Dataset collection from THS
* Retraining R^2 SAC paper, with same env still failed (TCN ACC: 0.708, RECALL: 0.183), will check with Han this week
* Paper reading and some plan (report this Fri)
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
* Train text enroll KWS model with aishell and kespeech data.
* Prepare live broadcast
||
*
||
*
|-

|-
|Lily
||
* Prepare for holiday course（October 2nd、3rd） and online-course
* AI radiance's daily work
||
*
||
*
|-

|-
|Turi
||
* Trained conformer on Sagalee data excluding utterances containing digits
** 2.65 % WER reduction (not as I expected)
* Preparing KWS data from Sagalee dataset using MFA
||
*
||
*
|-
|Yue Gu
||
* paper writing
* open the code
* prepare for the presentation
||
*
||
*
|-
|-
|Qi Qu
||
* KWS:
** Finding ideal thresholds for b0-models in predefined scenes: Chinese Mandarin, Cantonese, Uyghur and Kazakh.
** Finding ideal thresholds for b6-models with fixed b0-model thresholds.
* AED:
** Fixing parameters of Fbank feature extraction for CED and retraining classifiers.
||
*
||
*
|-

2024-09-09

2024-09-09T10:39:00Z

Turi：

{| class="wikitable"
!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
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
* mixed Hubert pretraining v1 result[https://z1et6d3xtb.feishu.cn/docx/O7cIdWjYZoVelYxJqqUcMjLmnxf](error)
** find a aligned problem(get a error aligned masked label)
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|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
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-09-02

2024-09-02T10:47:13Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
* Middle School AI book v0.0
||
*
||
*
|-

|-
|Lantian Li
||
* GPU status [https://z1et6d3xtb.feishu.cn/wiki/XGcGwRK5viJmpRkjH9AczIhynCh]
* Apply for the AP title (Failed)
* Submit two undergraduate thesis projects
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
* ICASSP exp and paper writing[https://z1et6d3xtb.feishu.cn/docx/MWO5dRL2FoGOpbx2IqRcDjgfnHh]
||
*
||
*
|-

|-
|Junming Yuan
||
* mixed Hubert pretraining v1 (in progress)
** still have some bugs
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*CNVSRC 2024 Website Finish
*ICASSP exp and paper writing[https://z1et6d3xtb.feishu.cn/docx/ZI2GdERFWoLGYNxDIfmcaPYxnwt?from=from_copylink]
||
*
||
*
|-

|-
|Pengqi Li
||
* [https://z1et6d3xtb.feishu.cn/docx/ZsWRd4qi8ocZEwx6oPbcZG2unvc]Pondering the expected conclusions for paper
* Experiment on timit(Finding and disadvantage)
** poor consistency of TAO and LayerCAM methods
** Toy Experiment
* Challenge: Do the conclusions drawn from SID tasks (toy experiments) align with those from SV tasks (more SOTA models)?
||
* To investigate and reproduce existing interpretability methods for verification task.
* To analyze the importance of phones using TTS datasets(broad coverage of phonemes) based SOTA models
||
*
|-

|-
|Wan Lin
||
* ICASSP paper writing
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*check the conditional chain code
||
*
||
*
|-

|-
|Junhui Chen
||
* Writing Neural Scoring paper, 1st ver. done.
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
* reproduce iTransformer
* Transfer iTrasnformer to financial data[https://z1et6d3xtb.feishu.cn/docx/ZwfLd73LZogGWFxtN6jcXAmxneh?from=from_copylink]
||
*
||
*
|-

|-
|Wenqiang Du
||
* Write middle school handbook(completed)
* Continue to training Chinese and Cantonese KWS model
||
*
||
*
|-

|-
|Yang Wei
||
* Get familiar with text enroll KWS training
||
*
||
*
|-

|-
|Lily
||
* AI-Radiance's daily work
||
*
||
*
|-

|-
|Turi
||
* Working on dataset paper refinement & additional experiment
** Attempting using pretrained model, not successful yet
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-

2024-08-26

2024-08-26T06:59:05Z

Turi：

{| class="wikitable"
!People !! This Week !! Next Week !! Task Tracking (DeadLine)
|-

|-
|Dong Wang
||
*
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Zhenghai You
||
*
||
*
||
*
|-

|-
|Junming Yuan
||
*
||
*
||
*
|-

|-
|Chen Chen
||
*
||
*
||
*
|-

|-
|Xiaolou Li
||
*
||
*
||
*
|-

|-
|Zehua Liu
||
*
||
*
||
*
|-

|-
|Pengqi Li
||
*
||
*
||
*
|-

|-
|Wan Lin
||
*
||
*
||
*
|-

|-
|Tianhao Wang
||
*
||
*
||
*
|-

|-
|Zhenyu Zhou
||
*
||
*
||
*
|-

|-
|Junhui Chen
||
*
||
*
||
*
|-

|-
|Jiaying Wang
||
*
||
*
||
*
|-

|-
|Yu Zhang
||
*
||
*
||
*
|-

|-
|Wenqiang Du
||
*
||
*
||
*
|-

|-
|Yang Wei
||
*
||
*
||
*
|-

|-
|Lily
||
*
||
*
||
*
|-

|-
|Turi
||
* Added more sections to the draft paper
** Need to refine and do more experiments
||
*
||
*
|-
|Yue Gu
||
*
||
*
||
*
|-
|-
|Qi Qu
||
*
||
*
||
*
|-