Difference between revisions of "0920 - Lantian Li"

From cslt Wiki
Jump to: navigation, search
(以“Weekly Summary 1. To go on studying a scoring method on GMM-UBM aiming to design a cohort reference speaker models. 1). Implement the K-means algorithem to cluster...”为内容创建页面)
 
 
Line 11: Line 11:
 
3). Re-score for the four parts on the cohort set.
 
3). Re-score for the four parts on the cohort set.
  
4). Score ranking for each part and draw score-rank distrubution diagram.  
+
4). Score ranking for each part and draw score-rank distrubution diagrams.  
  
 
Next Week
 
Next Week
  
1. Go on the task1 to explore the inherent law of re-scoring results and try to use the cohort set to  
+
1. Go on the task1 to explore the inherent law of re-scoring results and use the cohort set to  
  
 
reduce the error rate on the "Sensitive True Speaker"/"Sensitive Imp Speaker".
 
reduce the error rate on the "Sensitive True Speaker"/"Sensitive Imp Speaker".

Latest revision as of 10:22, 22 September 2014

Weekly Summary

1. To go on studying a scoring method on GMM-UBM aiming to design a cohort reference speaker models.

1). Implement the K-means algorithem to cluster the training set in order for organizing the cohort set.

2). Make the verfication score results dividied into four parts. --"Real True Speaker"/"Sensitive True

Speaker"/"Sensitive Imp Speaker"/"Abosulte Imp Speaker".

3). Re-score for the four parts on the cohort set.

4). Score ranking for each part and draw score-rank distrubution diagrams.

Next Week

1. Go on the task1 to explore the inherent law of re-scoring results and use the cohort set to

reduce the error rate on the "Sensitive True Speaker"/"Sensitive Imp Speaker".