Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS

Abstract: 

i-Vector feature representation with probabilistic linear discriminant analysis (PLDA) scoring in speaker recognition system has recently achieved effective performance even on channel mismatch conditions. In general, experiments carried out using this combined strategy employ linear discriminant analysis (LDA) after the i-Vector extraction phase to suppress irrelevant directions, such as those introduced by noise or channel distortions. However, speaker-related and -non-related variability present in the data may prevent LDA from finding the best projection matrix. In this study, we exclusively use support vectors of each class to find the optimum linear transformation. Post-processing of the i-Vectors by discriminant analysis via support vectors (SVDA) and traditional LDA is evaluated on NIST2010 speaker recognition evaluation (SRE) core and extended core (coreext) conditions. In addition, truncated coreext test data is used to examine the performance of the system for both long and short duration test segments. Computed equal error rate (EER) and minimum detection cost function (minDCF) criteria confirm consistent improvement of SVDA over traditional LDA. The relative improvement in terms of EER and minDCF with SVDA are about 32% and 9%, respectively.

up
0 users have voted:

Paper Details

Authors:
Fahimeh Bahmaninezhad, John H.L. Hansen
Submitted On:
25 March 2017 - 2:33am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Fahimeh Bahmaninezhad
Paper Code:
2244
Document Year:
2017
Cite

Document Files

Poster-Slides

(78 downloads)

Paper

(121 downloads)

Subscribe

[1] Fahimeh Bahmaninezhad, John H.L. Hansen, "I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1744. Accessed: Oct. 18, 2017.
@article{1744-17,
url = {http://sigport.org/1744},
author = {Fahimeh Bahmaninezhad; John H.L. Hansen },
publisher = {IEEE SigPort},
title = {I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS},
year = {2017} }
TY - EJOUR
T1 - I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS
AU - Fahimeh Bahmaninezhad; John H.L. Hansen
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1744
ER -
Fahimeh Bahmaninezhad, John H.L. Hansen. (2017). I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS. IEEE SigPort. http://sigport.org/1744
Fahimeh Bahmaninezhad, John H.L. Hansen, 2017. I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS. Available at: http://sigport.org/1744.
Fahimeh Bahmaninezhad, John H.L. Hansen. (2017). "I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS." Web.
1. Fahimeh Bahmaninezhad, John H.L. Hansen. I-VECTOR/PLDA SPEAKER RECOGNITION USING SUPPORT VECTORS WITH DISCRIMINANT ANALYSIS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1744