Portal del Investigador
Confidence based multiple classifier fusion in speaker verification
Abstract
A novel framework that applies Bayes-based confidence measure for multiple classifier system fusion is proposed. Compared with ordinary Bayesian fusion, the presented approach can lead to reductions as high as 37% and 35% in EER and ROC curve area, respectively, in speaker verification. © 2008 Elsevier B.V. All rights reserved.
Más información
| Título según WOS: | Confidence based multiple classifier fusion in speaker verification |
| Título según SCOPUS: | Confidence based multiple classifier fusion in speaker verification |
| Título de la Revista: | PATTERN RECOGNITION LETTERS |
| Volumen: | 29 |
| Número: | 7 |
| Editorial: | Elsevier |
| Fecha de publicación: | 2008 |
| Página de inicio: | 957 |
| Página final: | 966 |
| Idioma: | English |
| URL: | http://linkinghub.elsevier.com/retrieve/pii/S0167865508000378 |
| DOI: |
10.1016/j.patrec.2008.01.015 |
| Notas: | ISI, SCOPUS |