Cslt：/* Denoising & Farfield ASR */

2014-04-25T02:12:13Z

‎Denoising & Farfield ASR

Cslt：以内容“==Resoruce Building== * Maxi onboard * Release management should be started: Zhiyong (+) * Blaster 0.1 & vivian 0.0 system release == Leftover questions== * Asymmetric...”创建新页面

2014-04-25T02:11:19Z

以内容“==Resoruce Building== * Maxi onboard * Release management should be started: Zhiyong (+) * Blaster 0.1 & vivian 0.0 system release == Leftover questions== * Asymmetric...”创建新页面

新页面

==Resoruce Building==
* Maxi onboard
* Release management should be started: Zhiyong (+)
* Blaster 0.1 & vivian 0.0 system release

== Leftover questions==
* Asymmetric window: Great improvement on training set(WER 34% to 24%), however the improvement is lost on test. Overfitting?
* Multi GPU training: Error encountered
* Multilanguage training
* Investigating LOUDS FST.
* CLG embedded decoder plus online compiler.
* DNN-GMM co-training

== AM development ==

=== Sparse DNN ===
* GA-based block sparsity (+)
:* Found a paper in 2000 with similar ideas.
:* Try to get a student working on high performance computing to do the optimization

===Noise training===
:* More experiments with no-noise (+)
:* More experiments with additional noise types (+)

===AMR compression re-training===

* Stream model deliver to wechat server (Mengyuan + Liuchao)

===GFbank===

* GFBank Sinovoice test on 1700 MPE (10.34-10.14)
* GFBank sinovoice 1700 MPE stream

===Multilingual ASR===
* all phone strategy baseline done
* Some strange behavior observed when fixing early-leyers [http://cslt.riit.tsinghua.edu.cn/cgi-bin/cvss/cvss_request.pl?account=wangd&step=view_request&cvssid=205 click here]

===Denoising & Farfield ASR===

* Baseline: close-talk model decode far-field speech: 92.65
* MPE iteration: 92.78
91.15
91.21
91.51
* Will test the result on the dev set

===VAD===

* VAD bug on smoothing approach was found.

===Scoring===
* A speaker identification system based on ivector was delivered
* Male/female identification based on UBM was delievered
* Phone-sequence based graph decoding was delivered

==Word to Vector==

* Dimension of low space varies from 10-100 done. Expand to 200 dimensions. Some strange behavior was found on w2v. Try on daily people data.
* Test multi-classification from 2-9. w2v done. Work on lda.
* Test on various w2v window-size n=3-15. Strange behavior at n=9. [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/c/cd/Tubiao.pdf click here]

==LM development==

===NN LM===

* Character-based NNLM (6700 chars, 7gram), 500M data training done.
:* Overlfow found. Code change done. Run into 6 iterations.
* Investigate MS RNN LM training

==QA==

===FST-based matching===
:* Word-based FST 1-2 seconds with 1600 patterns. Huilan's implementation <1 second.
:* THRAX toolkit for grammar to FST

* Investigate determinization of G embedding
:* Refer to Kaldi new code

@@ 第42行： / 第42行： @@
 *  Baseline:  close-talk model decode far-field speech: 92.65
-*  MPE iteration:  92.78
+*  MPE1:  92.78
-.15
+   MPE2:  91.15
-.21
+   MPE3:  91.21
-.51
+   MPE4:  91.51
 * Will test the result on the dev set
 ===VAD===

2014-04-25 - 版本历史

Cslt：/* Denoising & Farfield ASR */

Cslt：以内容“==Resoruce Building== * Maxi onboard * Release management should be started: Zhiyong (+) * Blaster 0.1 & vivian 0.0 system release == Leftover questions== * Asymmetric...”创建新页面