cslt Wiki - 用户贡献 [zh-cn]

News-20161008

2016-11-09T16:17:10Z

Zengxy：/* Xiangyu Zeng */

==Members left CSLT recently==

===Caixia Wang===

Special Visitor, left for University of Diplomacy

===Maoning Wang===

Special Visitor, left for University of Central Economics and Finance

===[http://lty.cslt.org/ Tianyi Luo]===

Research engineer, left for University of California, Santa Cruz

[[文件:10.pic.jpg|200px]]

===Xiangyu Zeng===

Visiting students, left for University of Columbia

[[文件:681685309681956521.jpg|300px]]

===Yiqiao Pan===

Visiting student, left for University of Sydney

===Qixin Wang===

Visiting student, left for University of Southern California

[[文件:WechatIMG1.jpeg|300px]]

==New members joined CSLT recently==

===Yang Feng===
Special visitor

NLP

===Yang Wang===
Student

Financial processing

===Xingliang Cheng===

Student

Speech processing

===Shiyue Zhang===
Visitor

NLP

===Jiyuan Zhang===
Visitor

NLP

===Ying Shi===
Visitor

Speech processing

===Shiyao LI===
Intern

NLP

===Yixang Chen===
Intern

Speech processing

===Aodong Li===
Visitor

NLP

===Jingyi Lin===

Visitor

Speech processing

News-20161008

2016-11-09T16:16:59Z

Zengxy：/* Xiangyu Zeng */

==Members left CSLT recently==

===Caixia Wang===

Special Visitor, left for University of Diplomacy

===Maoning Wang===

Special Visitor, left for University of Central Economics and Finance

===[http://lty.cslt.org/ Tianyi Luo]===

Research engineer, left for University of California, Santa Cruz

[[文件:10.pic.jpg|200px]]

===Xiangyu Zeng===

Visiting students, left for University of Columbia
[[文件:681685309681956521.jpg|300px]]

===Yiqiao Pan===

Visiting student, left for University of Sydney

===Qixin Wang===

Visiting student, left for University of Southern California

[[文件:WechatIMG1.jpeg|300px]]

==New members joined CSLT recently==

===Yang Feng===
Special visitor

NLP

===Yang Wang===
Student

Financial processing

===Xingliang Cheng===

Student

Speech processing

===Shiyue Zhang===
Visitor

NLP

===Jiyuan Zhang===
Visitor

NLP

===Ying Shi===
Visitor

Speech processing

===Shiyao LI===
Intern

NLP

===Yixang Chen===
Intern

Speech processing

===Aodong Li===
Visitor

NLP

===Jingyi Lin===

Visitor

Speech processing

News-20161008

2016-11-09T16:16:34Z

Zengxy：/* Xiangyu Zeng */

==Members left CSLT recently==

===Caixia Wang===

Special Visitor, left for University of Diplomacy

===Maoning Wang===

Special Visitor, left for University of Central Economics and Finance

===[http://lty.cslt.org/ Tianyi Luo]===

Research engineer, left for University of California, Santa Cruz

[[文件:10.pic.jpg|200px]]

===Xiangyu Zeng===

Visiting students, left for University of Columbia
[[文件:681685309681956521.jpg|200px]]

===Yiqiao Pan===

Visiting student, left for University of Sydney

===Qixin Wang===

Visiting student, left for University of Southern California

[[文件:WechatIMG1.jpeg|300px]]

==New members joined CSLT recently==

===Yang Feng===
Special visitor

NLP

===Yang Wang===
Student

Financial processing

===Xingliang Cheng===

Student

Speech processing

===Shiyue Zhang===
Visitor

NLP

===Jiyuan Zhang===
Visitor

NLP

===Ying Shi===
Visitor

Speech processing

===Shiyao LI===
Intern

NLP

===Yixang Chen===
Intern

Speech processing

===Aodong Li===
Visitor

NLP

===Jingyi Lin===

Visitor

Speech processing

News-20161008

2016-11-09T16:15:00Z

Zengxy：/* Xiangyu Zeng */

==Members left CSLT recently==

===Caixia Wang===

Special Visitor, left for University of Diplomacy

===Maoning Wang===

Special Visitor, left for University of Central Economics and Finance

===[http://lty.cslt.org/ Tianyi Luo]===

Research engineer, left for University of California, Santa Cruz

[[文件:10.pic.jpg|200px]]

===Xiangyu Zeng===

Visiting students, left for University of Columbia
[[文件:10.pic.jpg|200px]]

===Yiqiao Pan===

Visiting student, left for University of Sydney

===Qixin Wang===

Visiting student, left for University of Southern California

[[文件:WechatIMG1.jpeg|300px]]

==New members joined CSLT recently==

===Yang Feng===
Special visitor

NLP

===Yang Wang===
Student

Financial processing

===Xingliang Cheng===

Student

Speech processing

===Shiyue Zhang===
Visitor

NLP

===Jiyuan Zhang===
Visitor

NLP

===Ying Shi===
Visitor

Speech processing

===Shiyao LI===
Intern

NLP

===Yixang Chen===
Intern

Speech processing

===Aodong Li===
Visitor

NLP

===Jingyi Lin===

Visitor

Speech processing

文件:681685309681956521.jpg

2016-11-09T16:13:16Z

Zengxy：

Xiangyu Zeng 2016-05-30

2016-05-30T02:27:13Z

Zengxy：以“Last week: 1.completed my thesis and prepared ppt for defense This week: 1.completed my dissertation defense and other graduation processes”为内容创建页面

Last week:

1.completed my thesis and prepared ppt for defense

This week:

1.completed my dissertation defense and other graduation processes

2016-05-30

2016-05-30T02:24:39Z

Zengxy：

[[Mengyuan Zhao 2016-05-30]]

[[Maoning Wang 2016-05-30]]

[[Aodong Li 2016-05-30]]

[[Xuewei Zhang 2016-05-30]]

[[Zhiyuan Tang 2016-05-30]]

[[Qixin Wang 2016-05-30]]

[[Chao Xing 2016-05-30]]

[[Xiangyu Zeng 2016-05-30]]

Xiangyu Zeng 2016-05-23

2016-05-23T01:11:39Z

Zengxy：

Last week:

1.completed the experiments, figure and paper of Online Learning

2.almost complete my thesis

This week:

1.finish the thesis and prepare for the last check

2.revise some details of online learning

Xiangyu Zeng 2016-05-23

2016-05-23T01:11:26Z

Zengxy：以“Last week: 1.completed the experiments, figure and paper of Online Learning 2.almost complete my thesis This week: 1.finish the thesis and prepare for the last c...”为内容创建页面

2016-05-23

2016-05-23T01:09:06Z

Zengxy：

[[Lantian Li 2016-05-23]]

[[Maoning Wang 2016-05-23]]

[[Zhiyong Zhang 2016-05-23]]

[[Zhiyuan Tang 2016-05-23]]

[[Xuewei Zhang 2016-05-23]]

[[Mengyuan Zhao 2016-05-23]]

[[Chao Xing 2016-05-23]]

[[Aodong Li 2016-05-23]]

[[Yiqiao Pan 2016-05-23]]

[[Xiangyu Zeng 2016-05-23]]

Xiangyu Zeng 2016-05-16

2016-05-16T01:27:41Z

Zengxy：以“Last week: 1.completed the online training experiments with Adgrad,Adedelta,Adam,NG. And finished the figures present in the paper 2.write my thesis This week: 1...”为内容创建页面

Last week:

1.completed the online training experiments with Adgrad,Adedelta,Adam,NG. And finished the figures present in the paper

2.write my thesis

This week:

1.complete my thesis

2.complete the paper

2016-05-16

2016-05-16T01:24:12Z

Zengxy：

[[Jingyi Lin 2016-05-16]]

[[Qixin Wang 2016-05-16]]

[[Lantian Li 2016-05-16]]

[[Mengyuan Zhao 2016-05-16]]

[[Aodong Li 2016-05-16]]

[[Yang Wang 2016-05-16]]

[[Xuewei Zhang 2016-05-16]]

[[Zhiyuan Tang 2016-05-16]]

[[Chao Xing 2016-05-16]]

[[Xiangyu Zeng 2016-05-16]]

Xiangyu Zeng 2016-05-09

2016-05-09T00:54:46Z

Zengxy：以“Last week: 1.reproduced the experiments on WSJ data set, and solved the net.sum infinite problem. 2.began to write my graduation paper. This week: 1.complete 1 a...”为内容创建页面

Last week:

1.reproduced the experiments on WSJ data set, and solved the net.sum infinite problem.

2.began to write my graduation paper.

This week:

1.complete 1 and go on 2

2.revise the online training paper

2016-05-09

2016-05-09T00:52:52Z

Zengxy：

[[Mengyuan Zhao 2016-05-09]]

[[Lantian Li 2016-05-09]]

[[Yang Wang 2016-05-09]]

[[Xuewei Zhang 2016-05-09]]

[[Zhiyong Zhang 2016-05-09]]

[[Chao Xing 2016-05-09]]

[[Zhiyuan Tang 2016-05-09]]

[[Aodong Li 2016-05-09]]

[[jiyuan Zhang 2016-05-09]]

[[Yiqiao Pan 2016-05-09]]

[[Xiangyu Zeng 2016-05-09]]

Xiangyu Zeng 2016-05-02

2016-05-01T14:15:15Z

Zengxy：以“Last week: 1.repeat experiments on the wsj dataset, but met a problem that KALDI_ASSERT: at nnet-train-frmshuff-adammax-last:Eval:nnet-loss.cc:73, failed: KALDI_ISF...”为内容创建页面

Last week:

1.repeat experiments on the wsj dataset, but met a problem that KALDI_ASSERT: at nnet-train-frmshuff-adammax-last:Eval:nnet-loss.cc:73, failed: KALDI_ISFINITE(net_out.Sum()). I guess that is caused by copied data, and the net train the same utterance continuous.

2.revised the paper of online learning, and change the dataset settings.

Next week:

1.fix the bug of 1, and add NG experiments into paper.

2016-05-02

2016-05-01T14:05:07Z

Zengxy：

[[Tianyi Luo 2016-05-02]]

[[Qixin Wang 2016-05-02]]

[[Xiangyu Zeng 2016-05-02]]

Xiangyu Zeng 2016-04-25

2016-04-25T01:41:55Z

Zengxy：

Last week:

1.debug the code of natural gradient and made it converged, and the result of NG seems good.

------------------------------------------------------------------------------------------------------------------------------
model | clean | airport | babble | restaurant | street
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c | 8.13 | 29.86 | 29.23 | 31.04 | 30.35
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a | 7.46 | 12.19 | 16.47 | 18.34 | 21.44
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b | 7.79 | 14.58 | 10.36 | 15.37 | 18.62
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b-r | 7.33 | 12.20 | 11.27 | 9.77 | 14.93
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b-r-s | 7.48 | 13.46 | 12.34 | 13.65 | 10.49
------------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c | 8.47 | 36.06 | 35.28 | 37.83 | 35.95
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a | 8.02 | 13.88 | 18.28 | 20.16 | 23.17
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b | 8.09 | 15.77 | 11.56 | 15.96 | 19.88
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b-r | 6.82 | 11.84 | 10.70 | 10.22 | 15.78
------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b-r-s | 7.33 | 15.12 | 12.97 | 13.65 | 10.53
------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c | 7.58 | 34.22 | 33.23 | 35.26 | 34.73
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a | 7.56 | 11.86 | 15.78 | 17.73 | 21.59
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b | 7.29 | 14.01 | 9.98 | 14.09 | 18.07
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b-r | 6.70 | 11.46 | 10.05 | 9.50 | 14.93
------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b-r-s | 6.91 | 14.09 | 12.38 | 13.33 | 9.88
------------------------------------------------------------------------------------------------------------------

Next week:

1.Add the experiments of NG into the technical report

Xiangyu Zeng 2016-04-25

2016-04-25T01:19:11Z

Zengxy：以“Last week: 1.debug the code of natural gradient and made it converged, and the result of NG seems good. -----------------------------------------------------------...”为内容创建页面

Last week:

1.debug the code of natural gradient and made it converged, and the result of NG seems good.

------------------------------------------------------------------------------------------------------------------------------
model | clean | airport | babble | restaurant | street
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c | 8.13 | 29.86 | 29.23 | 31.04 | 30.35
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a | 7.46 | 12.19 | 16.47 | 18.34 | 21.44
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b | 7.79 | 14.58 | 10.36 | 15.37 | 18.62
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b-r | 7.33 | 12.20 | 11.27 | 9.77 | 14.93
------------------------------------------------------------------------------------------------------------------------------
adam-max-lr-0.008-c-a-b-r-s | 7.48 | 13.46 | 12.34 | 13.65 | 10.49
------------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c | 8.47 | 36.06 | 35.28 | 37.83 | 35.95
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a | 8.02 | 13.88 | 18.28 | 20.16 | 23.17
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b | 8.09 | 15.77 | 11.56 | 15.96 | 19.88
----------------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b-r | 6.82 | 11.84 | 10.70 | 10.22 | 15.78
------------------------------------------------------------------------------------------------------------------
sgd-lr-0.008-c-a-b-r-s | 7.33 | 15.12 | 12.97 | 13.65 | 10.53
------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c | 7.58 | 34.22 | 33.23 | 35.26 | 34.73
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a | 7.56 | 11.86 | 15.78 | 17.73 | 21.59
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b | 7.29 | 14.01 | 9.98 | 14.09 | 18.07
----------------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b-r | 6.82 | 11.84 | 10.70 | 10.22 | 15.78
------------------------------------------------------------------------------------------------------------------
ng-lr-0.008-c-a-b-r-s | 6.91 | 14.09 | 12.38 | 13.33 | 9.88
------------------------------------------------------------------------------------------------------------------

2016-04-25

2016-04-25T01:10:11Z

Zengxy：

[[Tianyi Luo 2016-04-25]]

[[Mengyuan Zhao 2016-04-25]]

[[Zhiyuan Tang 2016-04-25]]

[[Hui Tang 2016-04-25]]

[[Yang Wang 2016-04-25]]

[[Xuewei Zhang 2016-04-25]]

[[Jiyuan Zhang 2016-04-25]]

[[Zhiyong Zhang 2016-04-25]]

[[Lantian Li 2016-04-25]]

[[Qixin Wang 2016-04-25]]

[[Chao Xing 2016-04-25]]

[[Xiangyu Zeng 2016-04-25]]

Xiangyu Zeng 2016-04-18

2016-04-18T00:11:01Z

Zengxy：以“Last week: 1.completed the code of NG, and try to apply it on the online training, but it diverged, so I'm fixing the bugs. 2.finished my middle check of graduate...”为内容创建页面

Last week:

1.completed the code of NG, and try to apply it on the online training, but it diverged, so I'm fixing the bugs.

2.finished my middle check of graduate design.

This week:

1.try best to make NG work well, and finish its online training work.

2.add experiments to the technical report of online training.

2016-04-18

2016-04-18T00:07:25Z

Zengxy：

[[Tianyi Luo 2016-04-18]]

[[Hui Tang 2016-04-18]]

[[Mengyuan Zhao 2016-04-18]]

[[Xiangyu Zeng 2016-04-18]]

Xiangyu Zeng 2016-04-11

2016-04-11T03:27:52Z

Zengxy：以“Last week: 1.read paper of Natural Gradient and began realizing the code 2.prepared my graduate middle check This week: 1.try to finish the code of NG and begin...”为内容创建页面

Last week:

1.read paper of Natural Gradient and began realizing the code

2.prepared my graduate middle check

This week:

1.try to finish the code of NG and begin some experiments of NG

2.read paper of lichun

2016-04-11

2016-04-11T03:26:01Z

Zengxy：

[[Mengyuan Zhao 2016-04-11]]

[[Lantian Li 2016-04-11]]

[[Hui Tang 2016-04-11]]

[[Chao Xing 2016-04-11]]

[[Yang Wang 2016-04-11]]

[[Zhiyong Zhang 2016-04-11]]

[[XueweiZhang 2016-04-11]]

[[Zhiyuan Tang 2016-04-11]]

[[Qixin Wang 2016-04-11]]

[[Xiangyu Zeng 2016-04-11]]

Xiangyu Zeng 2016-04-04

2016-04-04T10:46:46Z

Zengxy：以“Last week: 1.submitted the paper of Interspeech'16 2.began the next step of Online training, tried to realize some other learning rate strategies. This week: 1.g...”为内容创建页面

Last week:

1.submitted the paper of Interspeech'16

2.began the next step of Online training, tried to realize some other learning rate strategies.

This week:

1.go on 2

2.begin my graduation project

2016-04-04

2016-04-04T10:43:31Z

Zengxy：

[[Tianyi Luo 2016-04-04]]

[[Qixin Wang 2016-04-04]]

[[Hui Tang 2016-04-04]]

[[Chao Xing 2016-04-04]]

[[Xiangyu Zeng 2016-04-04]]

Xiangyu Zeng 2016-03-28

2016-03-28T00:50:58Z

Zengxy：以“Last week: 1.attended ICASSP 2.go on experiments on wsj corpus and almost complete This week: 1.Interspeech !!!”为内容创建页面

Last week:

1.attended ICASSP

2.go on experiments on wsj corpus and almost complete

This week:

1.Interspeech !!!

2016-03-28

2016-03-28T00:49:40Z

Zengxy：

[[Tianyi Luo 2016-03-28]]

[[Qixin Wang 2016-03-28]]

[[Hui Tang 2016-03-28]]

[[Zhiyong Zhang 2016-03-28]]

[[Yang Wang 2016-03-28]]

[[Chao Xing 2016-03-28]]

[[Xiangyu Zeng 2016-03-28]]

Xiangyu Zeng 2016-03-21

2016-03-21T00:58:36Z

Zengxy：以“Last week: 1.finished the Adamax and Sgd online traning with a initial good model. 2.Completed the first version paper of Interspeech. This week: 1.go on the 2 a...”为内容创建页面

Last week:

1.finished the Adamax and Sgd online traning with a initial good model.

2.Completed the first version paper of Interspeech.

This week:

1.go on the 2 and read more relevant paper to revise paper.

2016-03-21

2016-03-21T00:56:11Z

Zengxy：

[[Mengyuan Zhao 2016-03-21]]

[[Hui Tang 2016-03-21]]

[[Zhiyong Zhang 2016-03-21]]

[[Yiqiao Pan 2016-03-21]]

[[Chao Xing 2016-03-21]]

[[Yang Wang 2016-03-21]]

[[Zhang Xuewei 2016-03-21]]

[[Dongxu Zhang 2016-03-21]]

[[Zhiyuan Tang 2016-03-21]]

[[Qixin Wang 2016-03-21]]

[[Xiangyu Zeng 2016-03-21]]

Xiangyu Zeng 2016-03-14

2016-03-14T01:02:22Z

Zengxy：以“Last week: 1.finished the Adamax online learning based on the WSJ data set. 2.began to run the new experiments initialed with a converged model 3.began to write t...”为内容创建页面

Last week:

1.finished the Adamax online learning based on the WSJ data set.

2.began to run the new experiments initialed with a converged model

3.began to write the paper for interspeech

This week:

1.go on 2 and 3.

2016-03-14

2016-03-14T01:00:06Z

Zengxy：

[[Dongxu Zhang 2016-03-14]]

[[Zhiyuan Tang 2016-03-14]]

[[Hui Tang 2016-03-14]]

[[Tianyi Luo 2016-03-14]]

[[Chao Xing 2016-03-14]]

[[Mengyuan Zhao 2016-03-14]]

[[Yang Wang 2016-03-14]]

[[Xuewei Zhang 2016-03-14]]

[[Lantian Li 2016-03-14]]

[[Yiqiao Pan 2016-03-14]]

[[Xiangyu Zeng 2016-03-14]]

Xiangyu Zeng 2016-03-07

2016-03-07T01:42:56Z

Zengxy：以“Last week: 1.complemented the technique report of Adamax and updated it. 2.began to reproduce the experiments on a bigger data set, wsj, in order to illustrate som...”为内容创建页面

Last week:

1.complemented the technique report of Adamax and updated it.

2.began to reproduce the experiments on a bigger data set, wsj, in order to illustrate some details more clearly.

This week:

1.go on the (2) of last week.

2.begin to write the paper for interspeech

2016-03-07

2016-03-07T01:38:50Z

Zengxy：

[[Tianyi Luo 2016-03-07]]

[[Mengyuan Zhao 2016-03-07]]

[[Hui Tang 2016-03-07]]

[[Yang Wang 2016-03-07]]

[[Zhiyong Zhang 2016-03-07]]

[[Dongxu Zhang 2016-03-07]]

[[Xuewei Zhang 2016-03-07]]

[[Lantian Li 2016-03-07]]

[[Zhiyuan Tang 2016-03-07]]

[[Qixin Wang 2016-03-07]]

[[Yiqiao Pan 2016-03-07]]

[[Xiangyu Zeng 2016-03-07]]

Xiangyu Zeng 2016-02-01

2016-02-02T01:59:36Z

Zengxy：以“Last Week: 1.finished 1 copy, 5 copies, 10 copies training and paint pictures for them, and the results show a good performance of Adamax. 2.finished 1 hour, 2 hou...”为内容创建页面

Last Week:

1.finished 1 copy, 5 copies, 10 copies training and paint pictures for them, and the results show a good performance of Adamax.

2.finished 1 hour, 2 hours and 3 hours training set(divided the clean set into several parts and add noise into different parts to ensure the data won't be repeated) and the performance of Adamax can be represented better, the pictures are also painted to illustrate that.

Next Week:

1.revised the technical report and add the new experiments into it.

2.have spring festival, happy new year!

2016-02-01

2016-02-02T01:52:00Z

Zengxy：

[[Qixin Wang 2016-02-01]]

[[Mengyuan Zhao 2016-02-01]]

[[Chao Xing 2016-02-01]]

[[Xiangyu Zeng 2016-02-01]]

Xiangyu Zeng 2016-01-25

2016-01-25T09:50:42Z

Zengxy：以“Last week: 1.finished the 10 copies Adamax online training, and the results are followed our expectation. 2.tried some SNR and confirmed the noise level in 15db an...”为内容创建页面

Last week:

1.finished the 10 copies Adamax online training, and the results are followed our expectation.

2.tried some SNR and confirmed the noise level in 15db and snr 10db showing best.

Next week:

1.completed the 1 copy Adamax online training and reproduce the 5 copies one with new SNR to see if the results can show better.

2.draw some pictures of the results to illustrate our theory better.

2016-01-25

2016-01-25T09:46:55Z

Zengxy：

[[Qixin Wang 2016-01-25]]

[[Tianyi Luo 2016-01-25]]

[[Mengyuan Zhao 2016-01-25]]

[[Hui Tang 2016-01-25]]

[[Chao Xing 2016-01-25]]

[[Zhiyuan Tang 2016-01-25]]

[[Dongxu Zhang 2016-01-25]]

[[Xiaoxi Wang 2016-01-25]]

[[Xuewei Zhang 2016-01-25]]

[[Lantian Li 2016-01-25]]

[[Zhiyong Zhang 2016-01-25]]

[[Xiangyu Zeng 2016-01-25]]

Xiangyu Zeng 2016-01-18

2016-01-17T13:58:26Z

Zengxy：以“Last week: 1.prepared the training data for 10 copies Adamax online training and began experiments 2.made some comparisons between several SNR and finally chose no...”为内容创建页面

Last week:

1.prepared the training data for 10 copies Adamax online training and began experiments

2.made some comparisons between several SNR and finally chose noise level=15 db and sigma=10.

This week:

1.go on the experiments

2.have something personal things to do so call a week off

2016-01-18

2016-01-17T13:52:19Z

Zengxy：

[[Tianyi Luo 2016-01-18]]

[[Qixin Wang 2016-01-18]]

[[Mengyuan Zhao 2016-01-18]]

[[Zhiyong Zhang 2016-01-18]]

[[Zhiyuan Tang 2016-01-18]]

[[Xiangyu Zeng 2016-01-18]]

Xiangyu Zeng 2016-01-11

2016-01-12T06:48:19Z

Zengxy：以“Last Week: 1.completed the additional experiments of Adamax TRP and submitted the TRP. This Week: 1.go on with the experiments of Adamax in 10 copies situation”为内容创建页面

Last Week:

1.completed the additional experiments of Adamax TRP and submitted the TRP.

This Week:

1.go on with the experiments of Adamax in 10 copies situation

2016-01-11

2016-01-12T06:46:20Z

Zengxy：

[[Tianyi Luo 2016-01-11]]

[[Lantian Li 2016-01-11]]

[[Mengyuan Zhao 2016-01-11]]

[[Xiaoxi Wang 2016-01-11]]

[[Chao Xing 2016-01-11]]

[[Hui Tang 2016-01-11]]

[[Xiangyu Zeng 2016-01-11]]

Publication-trp

2016-01-07T07:51:10Z

Zengxy：

[[文件:lowv.png|200px]]
*[[媒体文件:Lowv.pdf|TRP-20150033: Learning Ordered Word Representations, Xiaoxi Wang, Chao Xing, Dong Wang, Rong Liu and Yiqiao Pan]]

[[文件:Adamax.png|200px]]
*[[媒体文件:Adamax Online Training for Speech Recognition.pdf|TRP-20150032: Adamax Online Training for Speech Recognition, Xiangyu Zeng, Zhiyong Zhang, Dong Wang]]

[[文件:Ptrnets.png|200px]]
*[[媒体文件: Ptrnets.pdf|TRP-20150031: An implementation of Pointer-Networks with Extensions, Xiaoxi Wang, Dong Wang]]

[[文件:dvad.png|200px]]
*[[媒体文件:dvad.pdf|TRP-20150030: DNN-based Voice Activity Detection for Speaker Recognition, Fanhu Bie, Zhiyong Zhang, Dong Wang, Thomas Fang Zheng]]

[[文件:uyghur.jpg|200px]]
*[[媒体文件:urghur.pdf|TRP-20150029: THUYG-20：A Free Uyghur Speech Database, Askar Rozi, Shi Yin, Zhiyong Zhang, Dong Wang, Askar Hamdulla]]

[[文件:nnpre.jpg|200px]]
*[[媒体文件:nnpre.pdf|TRP-20150028: Knowledge Transfer Pre-training, Zhiyuang Tang, Dong Wang]]

[[文件:mmargin.png|200px]]
*[[媒体文件:mmargin.pdf|TRP-20150027: Max-Margin Metric Learning for Speaker Recognition, Lantian Li, Chao Xing, Dong Wang, Thomas Fang Zheng]]

[[文件:binary.jpg|200px]]
*[[媒体文件:binary.pdf|TRP-20150026: Binary Speaker Embedding, Lantian Li, Chao Xing, Dong Wang, Kaimin Yu, Thomas Fang Zheng]]

[[文件:rnnrl.png|200px]]
*[[媒体文件:rnnrl.pdf|TRP-20150025: Relation Classification via Recurrent Neural Network, Dong Xu Zhang, Dong Wang, Rong Liu]]

[[文件:dplda.png|200px]]
*[[媒体文件:dplda.pdf|TRP-20150024: Learning from LDA using Deep Neural Networks, Dongxu Zhang, Tianyi Luo and Dong Wang]]

[[文件:jsrl.png|200px]]
*[[媒体文件:jsrl.pdf|TRP-20150023: Joint Semantic Relevance Learning with Text Data and Graph Knowledge, Dongxu Zhang, Bin Yuan, Dong Wang, Rong Liu]]

[[文件:listnet.png|200px]]
*[[媒体文件:listnet.pdf|TRP-20150022: Stochastic Top-k ListNet, Tianyi Luo, Dong Wang, Rong Liu, Yiqiao Pan]]

[[文件:segvector.png|200px]]
*[[媒体文件:segvector.pdf|TRP-20150021: Improved Deep Speaker Feature Learning for Text-Dependent Speaker Recognition, Lantian Li, Yiye Lin, Zhiyong Zhang, Dong Wang]]

[[文件:Vmclass.png|200px]]
*[[媒体文件:Vmclass.pdf|TRP-20150020: Document Classification with Spherical Word Vectors, Yiqiao Pan, Chao Xing, Dong Wang]]

[[文件:Tlearn.png|200px]]
*[[媒体文件:Tlearn.pdf|TRP-20150019: Transfer Learning for Speech and Language Processing, Dong Wang and Thomas Fang Zheng]]

[[文件:songci.png|200px]]
*[[媒体文件:songci.pdf|TRP-20150018: Chinese Song Iambics Generation with Neural Attention-based Model, Qixin Wang, Tianyi Luo, Dong Wang, Chao Xing]]

[[文件:database.jpg|200px]]
*[[媒体文件:Thuyg20-sre.pdf|TRP-20150017: AN OPEN/FREE DATABASE AND BENCHMARK FOR UYGHUR SPEAKER RECOGNITION, Askar Rozi, Dong Wang, Zhiyong Zhang, Thomas Fang Zheng]]

[[文件:Thchs.png|200px]]
*[[媒体文件:Thchs30.pdf|TRP-20150016: THCHS-30 : A Free Chinese Speech Corpus, Dong Wang and Xuewei Zhang]]

[[文件:Su.jpg|200px]]
*[[媒体文件:SUSR.pdf|TRP-20150015: Improving Short Utterance Speaker Recognition by Modeling Speech Unit Classes, Chenhao Zhang Dong Wang, Lantian Li and Thomas Fang Zheng]]

[[文件:dv.png|200px]]
*[[媒体文件:Dvector.pdf|TRP-20150014: Deep Speaker Vectors for Semi Text-independent Speaker Verification, Lantian Li, Dong Wang, Zhiyong Zhang and Thomas Fang Zheng]]

[[文件:dark.png|200px]]
*[[媒体文件:Dark.pdf|TRP-20150013: Recurrent Neural Network Training with Dark Knowledge Transfer, Dong Wang, Chao Liu, Zhiyuan Tang, Zhiyong Zhang, Mengyuan Zhao]]

[[文件:PBE.png|200px]]
*[[媒体文件:Probabilistic_Belief_Embedding_for_Knowledge_Population_(TRP).pdf|TRP-20150012: Probabilistic Belief Embedding for Large-scale Knowledge Population. Miao Fan, Qiang Zhou, Andrew Abel, Thomas Fang Zheng and Ralph Grishman]]

[[文件:fst-fw.png|200px]]
*[[媒体文件:wpair.pdf|TRP-20150011: Recognize Foreign Low-Frequency Words with Similar Pairs, Xi Ma1, Xiaoxi Wang1 and Dong Wang]]

[[文件:Cdae.png|200px]]
*[[媒体文件:Music.pdf|TRP-20150010: Music Removal by Denoising Autoencoder in Speech Recognition. Mengyuan Zhao, Dong Wang, Zhiyong Zhang and Xuewei Zhang]]

[[文件:vmfsne.png|200px]]
*[[媒体文件:Cslt-trp-template-vmfsne.pdf|TRP-20150009: VMF-SNE: Embedding for Spherical Data. Mian Wang, Dong Wang]]

[[文件:ros.png|200px]]
*[[媒体文件:Ros.pdf|TRP-20150008: Learning Speech Rate in Speech Recognition. Xiangyu Zeng, Shi Yin.]]

[[文件:Dnnvadstru.png|200px]]
*[[媒体文件:DNNVADTRP.pdf|TRP-20150007: Voice Activity Detection Based on Deep Neural Networks. Shi Yin.]] ([[媒体文件:Vad.pdf|Paper submiited to Tsinghua Xuebao]])

[[文件:Uyghur-training.png|200px]]
*[[媒体文件:UyghurTRP.pdf|TRP-20150006: Low-resource Uyghur Acoustic Model Training based on Cross-lingual Features. Shi Yin.]]

[[文件:Beam-forming.png|200px]]
*[[媒体文件:Multi-Microphones_Reverberation_Cancellation_for_Distant_Speech_Recognition.pdf|TRP-20150005: Multi-Microphones_Reverberation_Cancellation_for_Distant_Speech_Recognition. Xuewei Zhang.]]

[[文件:Clipping-speaker.png|200px]]
*[[媒体文件:Clip.pdf|TRP-20150004: Detection and Reconstruction of Clipped Speech in Speaker Recognition. Fanhu Bie et al.]]

[[文件:Semi-dynamic-embedding.png|200px]]
*[[媒体文件:Taglm.pdf|TRP-20150003: Semi-Dynamic Graph Embedding for Large Scale Language Model Adaptation. Bin Yuan et al.]]

[[文件:Speaker-discriminative-score.png|200px]]
*[[媒体文件:DNN-based Discriminative Scoring for Speaker.pdf|TRP-20150002: DNN-based Discriminative Scoring for Speaker Recognition Based on i-vector. Jun Wang et al. ]]

[[文件:Noisy-traiing.png|200px]]
*[[媒体文件:Noisy Training for Deep Neural Networks in.pdf|TRP-20150001: Noisy Training for Deep Neural Networks in Speech Recognition. Shi Yin et al.]]

[[文件:English-scroing.png|200px]]
*[[媒体文件:AutomaticScoringforEnglishUtterances.pdf|TRP-20140001: Automatic Scoring for English Utterances. Bo Hu.]]

[[媒体文件:Template.rar|CSLT-TRP latex template]]

文件:Adamax Online Training for Speech Recognition.pdf

2016-01-07T07:50:24Z

Zengxy：

Publication-trp

2016-01-07T07:47:26Z

Zengxy：

[[文件:lowv.png|200px]]
*[[媒体文件:Lowv.pdf|TRP-20150033: Learning Ordered Word Representations, Xiaoxi Wang, Chao Xing, Dong Wang, Rong Liu and Yiqiao Pan]]

[[文件:Adamax.png|200px]]
*[[媒体文件:Adamax.pdf|TRP-20150032: Adamax Online Training for Speech Recognition, Xiangyu Zeng, Zhiyong Zhang, Dong Wang]]

[[文件:Ptrnets.png|200px]]
*[[媒体文件: Ptrnets.pdf|TRP-20150031: An implementation of Pointer-Networks with Extensions, Xiaoxi Wang, Dong Wang]]

[[文件:dvad.png|200px]]
*[[媒体文件:dvad.pdf|TRP-20150030: DNN-based Voice Activity Detection for Speaker Recognition, Fanhu Bie, Zhiyong Zhang, Dong Wang, Thomas Fang Zheng]]

[[文件:uyghur.jpg|200px]]
*[[媒体文件:urghur.pdf|TRP-20150029: THUYG-20：A Free Uyghur Speech Database, Askar Rozi, Shi Yin, Zhiyong Zhang, Dong Wang, Askar Hamdulla]]

[[文件:nnpre.jpg|200px]]
*[[媒体文件:nnpre.pdf|TRP-20150028: Knowledge Transfer Pre-training, Zhiyuang Tang, Dong Wang]]

[[文件:mmargin.png|200px]]
*[[媒体文件:mmargin.pdf|TRP-20150027: Max-Margin Metric Learning for Speaker Recognition, Lantian Li, Chao Xing, Dong Wang, Thomas Fang Zheng]]

[[文件:binary.jpg|200px]]
*[[媒体文件:binary.pdf|TRP-20150026: Binary Speaker Embedding, Lantian Li, Chao Xing, Dong Wang, Kaimin Yu, Thomas Fang Zheng]]

[[文件:rnnrl.png|200px]]
*[[媒体文件:rnnrl.pdf|TRP-20150025: Relation Classification via Recurrent Neural Network, Dong Xu Zhang, Dong Wang, Rong Liu]]

[[文件:dplda.png|200px]]
*[[媒体文件:dplda.pdf|TRP-20150024: Learning from LDA using Deep Neural Networks, Dongxu Zhang, Tianyi Luo and Dong Wang]]

[[文件:jsrl.png|200px]]
*[[媒体文件:jsrl.pdf|TRP-20150023: Joint Semantic Relevance Learning with Text Data and Graph Knowledge, Dongxu Zhang, Bin Yuan, Dong Wang, Rong Liu]]

[[文件:listnet.png|200px]]
*[[媒体文件:listnet.pdf|TRP-20150022: Stochastic Top-k ListNet, Tianyi Luo, Dong Wang, Rong Liu, Yiqiao Pan]]

[[文件:segvector.png|200px]]
*[[媒体文件:segvector.pdf|TRP-20150021: Improved Deep Speaker Feature Learning for Text-Dependent Speaker Recognition, Lantian Li, Yiye Lin, Zhiyong Zhang, Dong Wang]]

[[文件:Vmclass.png|200px]]
*[[媒体文件:Vmclass.pdf|TRP-20150020: Document Classification with Spherical Word Vectors, Yiqiao Pan, Chao Xing, Dong Wang]]

[[文件:Tlearn.png|200px]]
*[[媒体文件:Tlearn.pdf|TRP-20150019: Transfer Learning for Speech and Language Processing, Dong Wang and Thomas Fang Zheng]]

[[文件:songci.png|200px]]
*[[媒体文件:songci.pdf|TRP-20150018: Chinese Song Iambics Generation with Neural Attention-based Model, Qixin Wang, Tianyi Luo, Dong Wang, Chao Xing]]

[[文件:database.jpg|200px]]
*[[媒体文件:Thuyg20-sre.pdf|TRP-20150017: AN OPEN/FREE DATABASE AND BENCHMARK FOR UYGHUR SPEAKER RECOGNITION, Askar Rozi, Dong Wang, Zhiyong Zhang, Thomas Fang Zheng]]

[[文件:Thchs.png|200px]]
*[[媒体文件:Thchs30.pdf|TRP-20150016: THCHS-30 : A Free Chinese Speech Corpus, Dong Wang and Xuewei Zhang]]

[[文件:Su.jpg|200px]]
*[[媒体文件:SUSR.pdf|TRP-20150015: Improving Short Utterance Speaker Recognition by Modeling Speech Unit Classes, Chenhao Zhang Dong Wang, Lantian Li and Thomas Fang Zheng]]

[[文件:dv.png|200px]]
*[[媒体文件:Dvector.pdf|TRP-20150014: Deep Speaker Vectors for Semi Text-independent Speaker Verification, Lantian Li, Dong Wang, Zhiyong Zhang and Thomas Fang Zheng]]

[[文件:dark.png|200px]]
*[[媒体文件:Dark.pdf|TRP-20150013: Recurrent Neural Network Training with Dark Knowledge Transfer, Dong Wang, Chao Liu, Zhiyuan Tang, Zhiyong Zhang, Mengyuan Zhao]]

[[文件:PBE.png|200px]]
*[[媒体文件:Probabilistic_Belief_Embedding_for_Knowledge_Population_(TRP).pdf|TRP-20150012: Probabilistic Belief Embedding for Large-scale Knowledge Population. Miao Fan, Qiang Zhou, Andrew Abel, Thomas Fang Zheng and Ralph Grishman]]

[[文件:fst-fw.png|200px]]
*[[媒体文件:wpair.pdf|TRP-20150011: Recognize Foreign Low-Frequency Words with Similar Pairs, Xi Ma1, Xiaoxi Wang1 and Dong Wang]]

[[文件:Cdae.png|200px]]
*[[媒体文件:Music.pdf|TRP-20150010: Music Removal by Denoising Autoencoder in Speech Recognition. Mengyuan Zhao, Dong Wang, Zhiyong Zhang and Xuewei Zhang]]

[[文件:vmfsne.png|200px]]
*[[媒体文件:Cslt-trp-template-vmfsne.pdf|TRP-20150009: VMF-SNE: Embedding for Spherical Data. Mian Wang, Dong Wang]]

[[文件:ros.png|200px]]
*[[媒体文件:Ros.pdf|TRP-20150008: Learning Speech Rate in Speech Recognition. Xiangyu Zeng, Shi Yin.]]

[[文件:Dnnvadstru.png|200px]]
*[[媒体文件:DNNVADTRP.pdf|TRP-20150007: Voice Activity Detection Based on Deep Neural Networks. Shi Yin.]] ([[媒体文件:Vad.pdf|Paper submiited to Tsinghua Xuebao]])

[[文件:Uyghur-training.png|200px]]
*[[媒体文件:UyghurTRP.pdf|TRP-20150006: Low-resource Uyghur Acoustic Model Training based on Cross-lingual Features. Shi Yin.]]

[[文件:Beam-forming.png|200px]]
*[[媒体文件:Multi-Microphones_Reverberation_Cancellation_for_Distant_Speech_Recognition.pdf|TRP-20150005: Multi-Microphones_Reverberation_Cancellation_for_Distant_Speech_Recognition. Xuewei Zhang.]]

[[文件:Clipping-speaker.png|200px]]
*[[媒体文件:Clip.pdf|TRP-20150004: Detection and Reconstruction of Clipped Speech in Speaker Recognition. Fanhu Bie et al.]]

[[文件:Semi-dynamic-embedding.png|200px]]
*[[媒体文件:Taglm.pdf|TRP-20150003: Semi-Dynamic Graph Embedding for Large Scale Language Model Adaptation. Bin Yuan et al.]]

[[文件:Speaker-discriminative-score.png|200px]]
*[[媒体文件:DNN-based Discriminative Scoring for Speaker.pdf|TRP-20150002: DNN-based Discriminative Scoring for Speaker Recognition Based on i-vector. Jun Wang et al. ]]

[[文件:Noisy-traiing.png|200px]]
*[[媒体文件:Noisy Training for Deep Neural Networks in.pdf|TRP-20150001: Noisy Training for Deep Neural Networks in Speech Recognition. Shi Yin et al.]]

[[文件:English-scroing.png|200px]]
*[[媒体文件:AutomaticScoringforEnglishUtterances.pdf|TRP-20140001: Automatic Scoring for English Utterances. Bo Hu.]]

[[媒体文件:Template.rar|CSLT-TRP latex template]]

Publication-trp

2016-01-07T07:43:35Z

Zengxy：

文件:Adamax.png

2016-01-07T07:40:58Z

Zengxy：

文件:AdaMax Online Training for Speech Recognition.pdf

2016-01-07T07:29:51Z

Zengxy：

Xiangyu Zeng 2016-01-04

2016-01-05T06:08:42Z

Zengxy：以“Last week: 1.finished the experiments of Adamax with street noise data, and didn't find a good result, then adopted the advise from teacher wang but still confused....”为内容创建页面

Last week:

1.finished the experiments of Adamax with street noise data, and didn't find a good result, then adopted the advise from teacher wang but still confused.

2.revised some mistakes on the last version of technique report.

This week:

1.try to fix the problem of street noise and find why it represent such characteristic.

2016-01-04

2016-01-05T06:04:57Z

Zengxy：

[[Tianyi Luo 2016-01-04]]

[[Qixin Wang 2016-01-04]]

[[Zhiyuan Tang 2016-01-04]]

[[Yiqiao Pan 2016-01-04]]

[[Xuewei Zhang 2016-01-04]]

[[Xiaoxi Wang 2016-01-04]]

[[Chao Xing 2016-01-04]]

[[Lantian Li 2016-01-04]]

[[Xiangyu Zeng 2016-01-04]]

Xiangyu Zeng 2015-12-28

2015-12-28T01:18:18Z

Zengxy：以“last week: 1.re-run the Adamax experiments and found some interesting things, I fixed some code problems and have some results in cvss, most are followed anticipati...”为内容创建页面

last week:

1.re-run the Adamax experiments and found some interesting things, I fixed some code problems and have some results in cvss, most are followed anticipation, but there are still some questions remain, I will find why.

This week:

finish the experiments and revise the technique report.