Difference between revisions of "2016-07-22"

From cslt Wiki
Jump to: navigation, search
(以“* Daily report # Use 4,6,8,10 as Gaussian mixture parameter to test the quality of voice conversion. ## Use 40 utterance to be the training set and the test set is...”为内容创建页面)
 
Line 1: Line 1:
 
* Daily report
 
* Daily report
  
# Use 4,6,8,10 as Gaussian mixture parameter to test the quality of voice conversion.
+
# Sibling speaker recognition.
## Use 40 utterance to be the training set and the test set is not come from the training set.
+
## Data preparation.
# Learn how to use Latex.
+
# Learn 'machine learning' courase.

Revision as of 04:21, 22 July 2016

  • Daily report
  1. Sibling speaker recognition.
    1. Data preparation.
  2. Learn 'machine learning' courase.