Difference between revisions of "Lantian Li 2016-08-22"

From cslt Wiki
Jump to: navigation, search
(以“* Weekly Report :* Deep speaker embedding (Random sample training). :* Deep speaker embedding ( LSTM + LR/Margin ) -- hold. :* OLE evaluation task. * This Wee...”为内容创建页面)
 
Line 1: Line 1:
 
* Weekly Report
 
* Weekly Report
 +
 +
:* OLE evaluation task. -- Done
  
 
:* Deep speaker embedding (Random sample training).
 
:* Deep speaker embedding (Random sample training).
  
 +
Make the d-vector baseline on Theano.
  
+
:* Local training.  
:* Deep speaker embedding ( LSTM + LR/Margin ) -- hold.
+
  
:* OLE evaluation task.
+
Data preparation -- Done.
 +
 
 +
T-matrix training.
 +
 
 +
:* Deep speaker embedding ( LSTM + LR/Margin ) -- hold.
  
 
* This Week
 
* This Week
  
 
:* Go on three tasks.
 
:* Go on three tasks.

Revision as of 03:37, 22 August 2016

  • Weekly Report
  • OLE evaluation task. -- Done
  • Deep speaker embedding (Random sample training).

Make the d-vector baseline on Theano.

  • Local training.

Data preparation -- Done.

T-matrix training.

  • Deep speaker embedding ( LSTM + LR/Margin ) -- hold.
  • This Week
  • Go on three tasks.