This paper presents an improved speaker verification technique that is especially appropriate for surveillance scenarios. The main idea is a meta-learning scheme aimed at improving fusion of low- and high-level speech information. While some existing systems fuse several classifier outputs, the proposed method uses a selective fusion scheme that takes into account conveying channel, speaking style and speaker stress as estimated on the test utterance. Moreover, we show that simultaneously employing multi-resolution versions of regular classifiers boosts fusion performance. The proposed selective fusion method aided by multi-resolution classifiers decreases error rate by 30% over ordinary fusion. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Solewicz, Y. A., & Koppel, M. (2005). Selective fusion for speaker verification in surveillance. In Lecture Notes in Computer Science (Vol. 3495, pp. 269–279). Springer Verlag. https://doi.org/10.1007/11427995_22
Mendeley helps you to discover research relevant for your work.