FreeNeb Status Report 2017-11-20 - 版本历史

2017年11月20日 (一) 01:59 Zhangzy

2017-11-20T01:59:11Z

2017年11月20日 (一) 01:56 Zhaomy

2017-11-20T01:56:52Z

2017年11月20日 (一) 01:45 Zhangshuai

2017-11-20T01:45:43Z

2017年11月20日 (一) 01:15 Zhaomy

2017-11-20T01:15:47Z

Zhaomy：以“This Week: {| class="wikitable" !People !! Last Week !! This Week !! Next Week !! Task Tracing(DeadLine) |- |Mengyuan Zhao || * Engineering...”为内容创建页面

2017-11-20T01:13:19Z

以“This Week: {| class="wikitable" !People !! Last Week !! This Week !! Next Week !! Task Tracing(DeadLine) |- |Mengyuan Zhao || * Engineering...”为内容创建页面

新页面

This Week:
{| class="wikitable"
!People !! Last Week !! This Week !! Next Week !! Task Tracing(DeadLine)
|-
|Mengyuan Zhao ||
* Engineering
# Draft API document of embedded-ASR engine.
# Draft API document of Deep Feature Extractor(deepfe).
# Draft a server version ASR DEMO, but still have bugs.
||
* Engineering
# Try to finish server version of TTS DEMO.
||

||

|-
|Zhiyong Zhang ||
* Train multi-speaker TTS based on Huilian and roobo data
:* Base model done, but the synthesised wav is not good. It seems the acoustic model does not converge.

||
* Continue to find the problem of poor acoustic predicting of Multi-speaker TTS;
* To train duration-model using 16k data.

||

||

|-
|Yang Wei ||
* Write test specification for FreeNeb TTS engine.
||
* Test FreeNeb TTS engine.
||

||

|-
|Dong Wang||
* ICASSP
||
* OC2017
||

||

|-
|Zhenlong Han||
* Finish training Japanese acoustic model with transfer learning. Now the MPE is on training.

ECO135 ECO_77 ETT SPC160 
baseline_4X1200X9391_xent 21.06 13.23 26.73 16.6 
embedded_6X512X800_xent 26.85 18.38 33.22 24.02 
embedded_6X400X800_transfer_learning_xent 23.8 15.71 29.66 19.64 

* VAD is finished.
||
* English embedded model training.
* Uyghur embedded model training.
||

||

|-
|Shuai Zhang||
* Add the intention of statistics into the graph and test it.
||
* According to the feedback， modify the project
* Complete the VVParrot project
* Add the documents about the two project
||

||

|-
|Yanchi Jin||
* Complete the megrez_tool (Fizzim) output format.
* Start the compilation of the unit testing and overall test framework.
||
* Complete the compilation of the framework.
||

||
|-
|}

@@ 第17行： / 第17行： @@
 |-
 |Zhiyong Zhang ||
-* Train multi-speaker TTS based on Huilian and roobo data
+* Fix code bug, and can reproduce the TTS experiment as Doc. Wang scripts based on huilian female1000
-:* Base model done, but the synthesised wav is not good. It seems the acoustic model does not converge.
+:* Large dataset test delayed due to the problems of grid
 ||
-* Continue to find the problem of poor acoustic predicting of Multi-speaker TTS;
+* Re-produce the Multi-speaker TTS;
 * To train duration-model using 16k data.

@@ 第51行： / 第51行： @@
 |-
 |Zhenlong Han||
-* Finish training Japanese acoustic model with transfer learning. Now the MPE is on training.
+* Finish training Japanese acoustic model MPE, but It's get worse along epoch increase. Mengyuan told me that I should tuning all parameters in the Nnet. I will try again with present degs.
+* Finish Uyghur embedded and cloud model xent training. But It is small worse than thuyg20 challenges, both Mengyuan and I dose not know more about the thuyg20 challenges model.
 ||
 * English embedded model training.

@@ 第63行： / 第63行： @@
 |Shuai Zhang||
 * Add the intention of statistics into the graph and test it.
 ||
 * According to the feedback， modify the project