DocumentCode
3065402
Title
Automatic speaker recognition based upon various distances of LSP frequencies
Author
Liu, Chi-Shi ; Huang, Chao-Shih ; Lin, Min-Tau ; Wang, Hsiao-Chuan
Author_Institution
Telecommun. Labs., Minist. of Commun., Taiwan
fYear
1991
fDate
1-3 Oct 1991
Firstpage
104
Lastpage
109
Abstract
A way to identify people by voice is discussed. A speaker´s spectral information represented by line spectrum pair (LSP) frequencies is used to describe characteristics of the speaker´s utterance and the VQ (vector quantization) method is used to model the spectral distribution of each speaker. Some easily computed distances (Euclidean distance, weighted distance by F-ratio, and hard-limited distance) are used to measure the discrimination among speakers
Keywords
speech recognition; vector quantisation; Euclidean distance; F-ratio; automatic speaker recognition; hard-limited distance; line spectrum pair frequencies; speaker discrimination; spectral distribution; utterance; vector quantization; weighted distance; Cepstrum; Chaotic communication; Euclidean distance; Frequency; Humans; Laboratories; Loudspeakers; Predictive models; Protection; Speaker recognition;
fLanguage
English
Publisher
ieee
Conference_Titel
Security Technology, 1991. Proceedings. 25th Annual 1991 IEEE International Carnahan Conference on
Conference_Location
Taipei
Print_ISBN
0-7803-0120-X
Type
conf
DOI
10.1109/CCST.1991.202199
Filename
202199
Link To Document