Open AccessFeature recovery for noise‐robust speaker verification
Author(s) -
Huang Houjun,
Xu Yunfei,
Zhou Ruohua,
Yan Yonghong
Publication year - 2015
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
ISSN - 1350-911X
DOI - 10.1049/el.2015.1418
Subject(s) - feature (linguistics) , speaker verification , computer science , feature vector , noise (video) , pattern recognition (psychology) , frame (networking) , artificial intelligence , speech recognition , speaker recognition , telecommunications , philosophy , linguistics , image (mathematics)
Noisy condition is an important extrinsic degradation affecting speaker verification system performance. A feature‐recovery approach is proposed to eliminate noise‐dependent variability in feature space. A frame of the noisy feature vector is recovered using the information of itself and the neighbour feature vectors. Experiments are conducted on noisy test sets for text‐dependent speaker verification tasks and the results indicate that the system can achieve significant performance improvement by using recovered feature vectors.