ASR Status Report 2018-1-2 - 版本历史

2018年1月8日 (一) 05:03 Tangzy

2018-01-08T05:03:00Z

2018年1月8日 (一) 04:23 Lilt

2018-01-08T04:23:02Z

2018年1月8日 (一) 04:21 Lilt

2018-01-08T04:21:57Z

2018年1月2日 (二) 06:17 Tangzy

2018-01-02T06:17:07Z

Tangzy：以“ {| class="wikitable" !Date!!People !! Last Week !! This Week !! Task Tracking |- | rowspan="8"|2018.1.2 |- |Miao Zhang || * || * || * |- |- |Ying Shi |...”为内容创建页面

2018-01-02T06:15:45Z

以“ {| class="wikitable" !Date!!People !! Last Week !! This Week !! Task Tracking |- | rowspan="8"|2018.1.2 |- |Miao Zhang || * || * || * |- |- |Ying Shi |...”为内容创建页面

新页面

{| class="wikitable"
!Date!!People !! Last Week !! This Week !! Task Tracking
|-
| rowspan="8"|2018.1.2

|-
|Miao Zhang
||
*
||
*
||
*
|-

|-
|Ying Shi
||
*
||
*
||
*
|-

|-
|Lantian Li
||
*
||
*
||
*
|-

|-
|Zhiyuan Tang
||
*
||
*
||
|-

|}

----

{| class="wikitable"
!Date!!People !! Last Week !! This Week !! Task Tracking
|-
| rowspan="8"|2017.12.25

|-
|Miao Zhang
||
*
||
* Read the 16k model script
* The cough recognition codes left by Xiaofei
||
* check the trivial database, make it more reasonable
* test the 16k model on the database
|-

|-
|Ying Shi
||
* some function for voice-printer
** speaker vector per utterance [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/6/63/SpkerVector2.png here]
** speaker vector minus base speaker vector [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/6/6b/Spkear_vector.png here]
* CTC for Haibo Wang (Token accuracy on train set 92.80%, on cv set 89.74%) haven't test on test set
* QRcode
** speaker vector merge phone grayscale [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/f/f3/Speaker_factor_gray.png here]
** speaker vector merge phone black-and-white map [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/9/97/1514176866%281%29.png here]
** speaker vector merge phone black-and-white map minus base vector [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/4/4e/SpeakerQrCode2.png here]
* ivector baseline for kazak-uyghur LRE performance is 81.85% (Utt level)
||
* Finish voice-checker copyright and submit the copyright in this Wednesday
||
*
|-

|-
|Lantian Li
||
* Complete the recipe for `VV_FACTOR`.
* 16K and 8K deep speaker model comparison.[http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=646]
||
* Patent for `VV_QuickMark`.
* Complete the demo for `VV_FACTOR`.[Assign to Shouyi Dai]
* Phonetic speaker embedding.
* Overlap training for speaker features.
||
*
|-

|-
|Zhiyuan Tang
||
* word level pronunciation accuracy based on likelihood (tell which word is well pronounced as '0' or badly pronounced '1')
||
* model adaptation
* if possible, an alpha version Parrot for test inside lab to collect some data for better configurature
||
|-

|}

@@ 第49行： / 第49行： @@
 |Zhiyuan Tang
 ||
+* A brief survey on oral language evaluation, both on application and algorithm
 ||
+* A test version of Parrot
 ||
+*

@@ 第34行： / 第34行： @@
 |Lantian Li
 ||
-* Commercial deep speaker model training in process. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=646]
+* Commercial deep speaker model training. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=646]
 * Phone-aware scoring on deep speaker feature. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=643]
-* Phonetic speaker embedding in process. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=644]
+* Phonetic speaker embedding. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=644]
 * Overlap training for speaker features. [http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=lilt&step=view_request&cvssid=645]
 ||
 * Commercial deep speaker model training.
 * Phone-aware scoring on deep speaker feature.
-* Phonetic speaker embedding in process.
+* Phonetic speaker embedding.
 * Overlap training for speaker features.
 ||