|
|
第33行: |
第33行: |
| * Continue joint training analysis work, but I'm very confused about how to improve | | * Continue joint training analysis work, but I'm very confused about how to improve |
| || | | || |
− | * Compare decode result between mono and bi LM | + | * Compare decode result between mono and bi LM, and the decode result ues bi LM before and after joint |
| * Choose wrong decode sentence and find its difference between baseline and shareGMM baseline | | * Choose wrong decode sentence and find its difference between baseline and shareGMM baseline |
| * Finished ML book | | * Finished ML book |
Date |
People |
Last Week |
This Week
|
2016.12.19
|
Jingyi Lin
|
|
- Concentrate on checking the cslt.book.
- Prepare for the annual convention.
|
Yanqing Wang
|
- build a data sender ( read & generate txt files of distracted feature )
|
- build a data analyzer ( detect the modification of files and make response ( show tokens ) )
|
Hang Luo
|
- Continue joint training analysis work, but I'm very confused about how to improve
|
- Compare decode result between mono and bi LM, and the decode result ues bi LM before and after joint
- Choose wrong decode sentence and find its difference between baseline and shareGMM baseline
- Finished ML book
|
Ying Shi
|
- some work about kazak lm
- crawl data from kazak internet
|
- run new AM by current speech data
- get more corpus from internet
- use current corpus make LM and decode
|
Yixiang Chen
|
- Leanring tensorflow
- coding pair wise net use tensorflow
- alter CNN
|
- coding CNN connect pair wise
- Dealing with the issue of different lengths of voice
|
Lantian Li
|
- LRE challenge on AP16-OL7.
- Jeju for APSIPA16.
|
- LRE on AP16-OL7.
- Deep speaker embedding.
|
Zhiyuan Tang
|
|
- A speech about recent ASR improvements.
- A supplementary TRP for "Multi-task Recurrent Model for True Multilingual Speech Recognition".
|
Date |
People |
Last Week |
This Week
|
2016.12.12
|
Yanqing Wang
|
- read a paper about driving distraction detection task
|
- show normal/distraction patterns of a driver with one class and two class SVM
|
Hang Luo
|
- Compared mono-language model and bi-language model decode result.
- Read paper of WFST.
|
- Use different corpus or generate mix-lingual corpus to run experiments
|
Ying Shi
|
- work from Chao Xing down
- kazak lm
- got some corpus from a student who study in Minzu University of China.But the corpus is short (about 10000) so the ppl is also poor.
- spider
|
|
Yixiang Chen
|
- Complete the replay task experiment and report
|
- learning tensorflow coding DNN and CNN net
|
Lantian Li
|
- interim report done;
- PPT for APSIPA16;
- LRE challenge on AP16-OL7.
- Deep speaker embedding restart!
- Submit TRP-20160011 on Replay detection.
|
|
Zhiyuan Tang
|
- interim report done;
- PPT for APSIPA16;
- language mask[1]
|
|