A proposed likelihood transformation for speaker verification

Authors:
D. Tran;M. Wagner
Affiliations:
Sch. of Comput., Canberra Univ., ACT, Australia;-
Venue:
ICASSP '00 Proceedings of the Acoustics, Speech, and Signal Processing, 2000. on IEEE International Conference - Volume 02
Year:
2000

Citing 0
Cited 3

Noise Clustering-Based Speaker Verification

AFSS '02 Proceedings of the 2002 AFSS International Conference on Fuzzy Systems. Calcutta: Advances in Soft Computing
Fuzzy C-Means Clustering-Based Speaker Verification

AFSS '02 Proceedings of the 2002 AFSS International Conference on Fuzzy Systems. Calcutta: Advances in Soft Computing
Utterance verification using fuzzy methods

ICECS'03 Proceedings of the 2nd WSEAS International Conference on Electronics, Control and Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Most of current normalisation methods for speaker verification are based on the ratio of the claimed speaker's and impostors' likelihood functions. As an extension, we propose alternative normalisation methods, which are based on the ratio of functions of the claimed speaker's and impostors' likelihood functions. A likelihood transformation is proposed in this paper to relate current and proposed methods. We also report several significant experimental results performed on the T146, the ANDOSL and the YOHO corpora using these proposed normalisation methods.