13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Discriminative Decision Function Based Scoring Method in Joint Factor Analysis for Speaker Verification

Chunyan Liang, Xiang Zhang, Lin Yang, Yonghong Yan

Key Laboratory of Speech Acoustics and Content Understanding, Institute of Acoustics, Chinese Academy of Sciences, China

This paper introduces a discriminative decision function scoring method for speaker recognition with the Joint Factor Analysis (JFA) system. In the scoring module of JFA system, an approximate form of the decision function is proposed. Based on the approximation, we present a discriminative decision function by re-estimating the contribution of each speech sound unit to the decision function. The discriminative decision function is used to exploit the individual Gaussian component for better classification. The experiments are carried out on the telephone-telephone core condition of NIST SRE2010. The experimental results show that the proposed scoring method outperforms the conventional frame-by-frame and integration scoring strategies, respectively providing up to 19% and 15% relative improvement.

Index Terms: speaker verification, joint Factor Analysis, discriminative decision function

Full Paper

Bibliographic reference.  Liang, Chunyan / Zhang, Xiang / Yang, Lin / Yan, Yonghong (2012): "Discriminative decision function based scoring method in joint factor analysis for speaker verification", In INTERSPEECH-2012, 1564-1567.