TY - GEN
T1 - Classifier subset selection and fusion for speaker verification
AU - Sedlák, Filip
AU - Kinnunen, Tomi
AU - Hautamäki, Ville
AU - Lee, Kong Aik
AU - Li, Haizhou
PY - 2011/5
Y1 - 2011/5
N2 - State-of-the-art speaker verification systems consists of a number of complementary subsystems whose outputs are fused, to arrive at more accurate and reliable verification decision. In speaker verification, fusion is typically implemented as a linear combination of the subsystem scores. Parameters of the linear model are commonly estimated using the logistic regression method, as implemented in the popular FoCal toolkit. In this paper, we study simultaneous use of classifier selection and fusion. We study four alternative fusion strategies, three score warping techniques, and provide interesting experimental bounds on optimal classifier subset selection. Detailed experiments are carried out on the NIST 2008 and 2010 SRE corpora.
AB - State-of-the-art speaker verification systems consists of a number of complementary subsystems whose outputs are fused, to arrive at more accurate and reliable verification decision. In speaker verification, fusion is typically implemented as a linear combination of the subsystem scores. Parameters of the linear model are commonly estimated using the logistic regression method, as implemented in the popular FoCal toolkit. In this paper, we study simultaneous use of classifier selection and fusion. We study four alternative fusion strategies, three score warping techniques, and provide interesting experimental bounds on optimal classifier subset selection. Detailed experiments are carried out on the NIST 2008 and 2010 SRE corpora.
KW - Classifier selection
KW - linear fusion
UR - http://www.scopus.com/inward/record.url?scp=80051625290&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2011.5947365
DO - 10.1109/ICASSP.2011.5947365
M3 - Conference article published in proceeding or book
AN - SCOPUS:80051625290
SN - 9781457705397
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 4544
EP - 4547
BT - 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011 - Proceedings
T2 - 36th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011
Y2 - 22 May 2011 through 27 May 2011
ER -