On the use of nearest feature line for speaker identification

doi:10.1016/S0167-8655(02)00147-2

CORC > 北京大学 > 信息科学技术学院

	On the use of nearest feature line for speaker identification
	Chen, K ; Wu, TY ; Zhang, HJ
刊名	pattern recognition letters
	2002
关键词	nearest feature line speaker identification dynamic time warping vector quantization nearest neighboring measure RECOGNITION CLASSIFICATION RETRIEVAL
DOI	10.1016/S0167-8655(02)00147-2
英文摘要	As a new pattern classification method, nearest feature line (NFL) provides an effective way to tackle the sort of pattern recognition problems where only limited data are available for training. In this paper, we explore the use of NFL for speaker identification in terms of limited data and examine how the NFL performs in such a vexing problem of various mismatches between training and test. In order to speed up NFL in decision-making, we propose an alternative method for similarity measure. We have applied the improved NFL to speaker identification of different operating modes. Its text-dependent performance is better than the dynamic time warping (DTW) on the Ti46 corpus, while its computational load is much lower than that of DTW. Moreover, we propose an utterance partitioning strategy used in the NFL for better performance. For the text-independent mode, we employ the NFL to be a new similarity measure in vector quantization (VQ), which causes the VQ to perform better on the KING corpus. Some computational issues on the NFL are also discussed in this paper. (C) 2002 Elsevier Science B.V. All rights reserved.; Computer Science, Artificial Intelligence; SCI(E); EI; 16; ARTICLE; 14; 1735-1746; 23
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/256562]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Chen, K,Wu, TY,Zhang, HJ. On the use of nearest feature line for speaker identification[J]. pattern recognition letters,2002.
APA	Chen, K,Wu, TY,&Zhang, HJ.(2002).On the use of nearest feature line for speaker identification.pattern recognition letters.
MLA	Chen, K,et al."On the use of nearest feature line for speaker identification".pattern recognition letters (2002).