TY - GEN
T1 - Maximizing distance between GMMs for speaker verification using particle swarm optimization
AU - Kim, Min Seok
AU - Yang, I. L.Ho
AU - Yu, Ha Jin
PY - 2008
Y1 - 2008
N2 - In this paper, we propose a feature transformation method to maximize the distances between the Gaussian mixture models for speaker verification. The feature transformation matrix is optimized by using particle swarm optimization. We evaluate the transformation using YOHO speech data, and the transformation is applied to some speakers who give poor performance. As the result, the overall equal error rate is reduced to 1.71% from 1.97% of the baseline.
AB - In this paper, we propose a feature transformation method to maximize the distances between the Gaussian mixture models for speaker verification. The feature transformation matrix is optimized by using particle swarm optimization. We evaluate the transformation using YOHO speech data, and the transformation is applied to some speakers who give poor performance. As the result, the overall equal error rate is reduced to 1.71% from 1.97% of the baseline.
UR - http://www.scopus.com/inward/record.url?scp=57649201695&partnerID=8YFLogxK
U2 - 10.1109/ICNC.2008.820
DO - 10.1109/ICNC.2008.820
M3 - Conference contribution
AN - SCOPUS:57649201695
SN - 9780769533049
T3 - Proceedings - 4th International Conference on Natural Computation, ICNC 2008
SP - 175
EP - 178
BT - Proceedings - 4th International Conference on Natural Computation, ICNC 2008
T2 - 4th International Conference on Natural Computation, ICNC 2008
Y2 - 18 October 2008 through 20 October 2008
ER -