Sinovoice-2014-04-15 - 版本历史

2014年4月15日 (二) 08:26 Cslt

2014-04-15T08:26:15Z

2014年4月15日 (二) 08:25 Cslt

2014-04-15T08:25:53Z

2014年4月15日 (二) 07:25 Cslt

2014-04-15T07:25:38Z

Cslt：以内容“=Environment setting= * Sinovoice internal server deployment. Now a better approach by using Gitlab + Redmine. * Delivered part of the Kaldi code. Some fix yet waitin...”创建新页面

2014-04-15T06:08:17Z

以内容“=Environment setting= * Sinovoice internal server deployment. Now a better approach by using Gitlab + Redmine. * Delivered part of the Kaldi code. Some fix yet waitin...”创建新页面

新页面

=Environment setting=

* Sinovoice internal server deployment. Now a better approach by using Gitlab + Redmine.
* Delivered part of the Kaldi code. Some fix yet waiting for check in.

=Corpora=

* 300h Guangxi telecom text transcription prepared. 150h before 18th, April.
* Now totally 1338h (470 + 346 + 105BJ mobile + 200 PICC + 108h HBTc + 109h New BJ mobile) telephone speech is ready.
* 16k 6000h data: 978h online data from DataTang + 656h online mobile data + 4300h recording data.

* Standard established for LM-speech-text labeling (speech data transcription for LM enhancement)
* Xiaona will prepare noise database. Start from telephone speech.

=Acoustic modeling=

==Telephone model training==

===1000h Training===

*Baseline: 8k states, 470+300 MPE4, 20.29
* Jietong phone, 200 hour seed, 10k states training:
:* Xent 5 iteration: 23.26
:* Xent 16 iteration: 22.90

* CSLT phone, 8k states training
:* MPE1: 20.60
:* MPE2: 20.37
:* MPE3: 20.37
:* MPE4: 20.37

==6000 hour 16k training==

===Training progress===

* 6000h/CSLT phone set training
:* Xent: 12.83
:* MPE1: 9.21
:* MPE2: 9.13

* 6000h/jt phone set phone set training
:* lattice done, MPE run 4 days

===Train Analysis===

* The Qihang model used a subset of the 6k data
:* 2500+950H+tang500h*+20131220, approximately 1700+2400 hours

* GMM training using this subset achieved 22.47%. Xiaoming's result is 16.1%.
:* Seems the database is still not very consistent
:* Xiaoming kicked off the job to reproduce the Qihang training using this subset

===Multilanguage Training===

* Prepare Chinglish data: Wang Dong provide info.
* Prepare shared DNN structure for multilingual training

===Noise robust feature===

* GFbank can be propagated to Sinovoice
* Liuchao will prepare fast computing code

=Language modeling=

==Training recipe transfer==
* Training process was delivered.
* Problems in encoding were solved.
* Initial CSLT LM buildup completed

==Domain specific atom-LM construction==

===Some potential problems===
* Unclear domain definition
* Using the same development set (8k transcription) is not very appropriate

===Text data filtering===

* VSM-based topic segmentation was delivered to Sinovoice, but the tool is highly inefficient.
* An enhanced toolkit was delivered.

=DNN Decoder=

==decoder optimization==

* Test computation cost of each step
:* beam 9/5000: netforward 65%
:* beam 13/7000: netforward 28%
* Sinovoice change in Kaldi delivered and ready to check-in
* Need to verify the speed of the CSLT engine

==Frame-skipping==
* Zhiyong & Liuchao will deliver the frame-skipping approach.

==BigLM optimization==
* Investigate BigLM retrieval optimization.

←上一版本		2014年4月15日 (二) 08:26的版本
第41行：		第41行：

	* 6000h/jt phone set phone set training		* 6000h/jt phone set phone set training
−	:* ~~ran~~ into MPE1.	+	:* running into MPE1.

←上一版本		2014年4月15日 (二) 08:25的版本
第41行：		第41行：

	* 6000h/jt phone set phone set training		* 6000h/jt phone set phone set training
−	:* MPE1:	+	:* ran into MPE1.

@@ 第3行： / 第3行： @@
 * Sinovoice internal server deployment. Now a better approach by using Gitlab + Redmine.
 * Delivered part of the Kaldi code. Some fix yet waiting for check in.
+* Email notification is problematic. Need obtain a smtp server
 =Corpora=
@@ 第10行： / 第10行： @@
 * Now totally 1338h (470 + 346 + 105BJ mobile + 200 PICC + 108h HBTc + 109h New BJ mobile) telephone speech is ready.
 * 16k 6000h data: 978h online data from DataTang + 656h online mobile data + 4300h recording data.
 * Standard established for LM-speech-text labeling (speech data transcription for LM enhancement)
 * Xiaona will prepare noise database. Start from telephone speech.
@@ 第22行： / 第21行： @@
 *Baseline: 8k states, 470+300 MPE4, 20.29
 * Jietong phone, 200 hour seed, 10k states training:
 :* Xent 16 iteration: 22.90
+:* MPE1 : 20.89
 * CSLT phone, 8k states training
@@ 第40行： / 第38行： @@
 :* MPE1: 9.21
 :* MPE2: 9.13
 * 6000h/jt phone set  phone set training
-:* lattice done, MPE run 4 days
+:* MPE1:
@@ 第56行： / 第55行： @@
 ===Multilanguage Training===
-* Prepare Chinglish data: Wang Dong provide info.
+* Prepare Chinglish data: contacted with a vendor for 1000 hour mobile recording. Will check how much we need
 * Prepare shared DNN structure for multilingual training
@@ 第62行： / 第63行： @@
 * GFbank can be propagated to Sinovoice
 * Liuchao will prepare fast computing code
@@ 第69行： / 第71行： @@
 * Training process was delivered.
 * Problems in encoding were solved.
-* Initial CSLT LM buildup completed
+* Initial CSLT LM buildup completed.
 ==Domain specific atom-LM construction==
@@ 第79行： / 第81行： @@
 ===Text data filtering===
 * VSM-based topic segmentation was delivered to Sinovoice, but the tool is highly inefficient.
 * An enhanced toolkit was delivered.
+* A telecom specific word list is ready, several stop words are ready
 =DNN Decoder=