Paper Info
Title
Utilization of unlabeled development data for speaker verification
Abstract
State-of-the-art speaker verification systems model speaker identity by mapping i-Vectors onto a probabilistic linear discriminant analysis (PLDA) space. Compared to other modeling approaches (such as cosine distance scoring), PLDA provides a more efficient mechanism to separate speaker information from other sources of undesired variabilities and offers superior speaker verification performance. Unfortunately, this efficiency is obtained at the cost of a required large corpus of labeled development data, which is too expensive/unrealistic in many cases. This study investigates a potential solution to resolve this challenge by effectively utilizing unlabeled development data with universal imposter clustering. The proposed method offers +21.9% and +34.6% relative gains versus the baseline system on two public available corpora, respectively. This significant improvement proves the effectiveness of the proposed method.
Year
DOI
Venue
2014
10.1109/SLT.2014.7078611
SLT
Keywords
Field
DocType
labeled development data corpus,relative gains,speaker verification systems,i-vector,i-vectors,probabilistic linear discriminant analysis,plda,undesired variability sources,universal imposter clustering,speaker verification,unlabeled development data utilization,speaker recognition,public available corpora,speaker information,clustering,baseline system,speaker identity modelling,probability
Probabilistic linear discriminant analysis,I vector,Speaker verification,Pattern recognition,Computer science,Cosine Distance,Speech recognition,Speaker recognition,Artificial intelligence,Speaker diarisation,Baseline system,Cluster analysis
Conference
ISSN
Citations 
PageRank 
2639-5479
2
0.38
References 
Authors
14
6
Name
Order
Citations
PageRank
Gang Liu11048.14
Chengzhu Yu2163.77
Navid Shokouhi3356.43
Abhinav Misra4173.81
Hua Xing551.47
John H. L. Hansen63215365.75