TY - GEN
T1 - A new algorithm for speaker identification using the Dempster-Shafer theory of evidence
AU - Naseem, Imran
AU - Deriche, Mohamed
PY - 2006
Y1 - 2006
N2 - In this paper, speaker identification using the Dempster-Shafer theory of evidence is discussed. The objective is to use the complementary information present from different classifiers to fuse the classification results into a single decision. Here, we use a decreasing function of the distance (of the classifiers) as our belief function. In the case of speaker identification, we show that a combined classifier based on the Dempster-Shafer theory outperforms the indidvidual LPCC and MFCC classifiers when used separately.
AB - In this paper, speaker identification using the Dempster-Shafer theory of evidence is discussed. The objective is to use the complementary information present from different classifiers to fuse the classification results into a single decision. Here, we use a decreasing function of the distance (of the classifiers) as our belief function. In the case of speaker identification, we show that a combined classifier based on the Dempster-Shafer theory outperforms the indidvidual LPCC and MFCC classifiers when used separately.
UR - https://www.scopus.com/pages/publications/79957522121
M3 - Conference contribution
AN - SCOPUS:79957522121
SN - 9781932415957
T3 - Proceedings of the 2006 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV'06
SP - 47
EP - 51
BT - Proceedings of the 2006 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV'06
ER -