Paper Info
Title
Towards Structured Approaches to Arbitrary Data Selection and Performance Prediction for Speaker Recognition
Abstract
We developed measures relating feature vector distributions to speaker recognition (SR) performances for performance prediction and potential arbitrary data selection for SR. We examined the measures of mutual information, kurtosis, correlation, and measures pertaining to intra- and inter-speaker variability. We applied the measures on feature vectors of phones to determine which measures gave good SR performance prediction of phones standalone and in combination. We found that mutual information had an -83.5% correlation with the Equal Error Rates (EERs) of each phone. Also, Pearson's correlation between the feature vectors of two phones had a -48.6% correlation with the relative EER improvement of the score-level combination of the phones. When implemented in our new data-selection scheme (which does not require a SR system to be run), the measures allowed us to select data with 2.13% overall EER improvement (on SRE08) over data selected via a brute-force approach, at a fifth of the computational costs.
Year
DOI
Venue
2009
10.1007/978-3-642-01793-3_53
ICB
Keywords
Field
DocType
relative eer improvement,towards structured approaches,good sr performance prediction,performance prediction,phones standalone,sr system,potential arbitrary data selection,mutual information,overall eer improvement,speaker recognition,feature vector distribution,feature vector,arbitrary data selection,relevance,redundancy
Feature vector,Pattern recognition,Computer science,Speech recognition,Speaker recognition,Redundancy (engineering),Phone,Correlation,Mutual information,Artificial intelligence,Performance prediction,Kurtosis
Conference
Volume
ISSN
Citations 
5558
0302-9743
0
PageRank 
References 
Authors
0.34
11
1
Name
Order
Citations
PageRank
Howard Lei11126.90