Feature recovery for noise‐robust speaker verification. Issue 18 (1st September 2015)

• Record Type: Journal Article
• Authors: Huang, Houjun; Xu, Yunfei; Zhou, Ruohua; Yan, Yonghong
• Is Part Of: Electronics letters. Volume 51:Issue 18(2015)
• Page Start: 1459
• Page End: 1461
• Publication Date: 2015-09-01
• Subjects: speaker recognition -- vectors -- interference suppression -- feature extraction; feature‐recovery approach -- noise‐dependent variability -- feature space -- noisy feature vector -- text‐dependent speaker verification tasks -- recovered feature vectors -- noise‐robust speaker verification; Electronics -- Periodicals; 621.381
• DOI: 10.1049/el.2015.1418
• Languages: English
• ISSNs: 0013-5194

Abstract

Abstract : 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.