Please use this identifier to cite or link to this item: http://hdl.handle.net/11452/28798
Title: Principal component based classification for text-independent speaker identification
Authors: Uludağ Üniversitesi/Mühendislik Fakültesi/Elektrik Elektronik Mühendisliği Bölümü.
Hanilçi, Cemal
Ertaş, Figen
S-4967-2016
AAH-4188-2021
35781455400
24724154500
Keywords: Computer science
Engineering
Classifiers
Identification (control systems)
Independent component analysis
Loudspeakers
Soft computing
Speech recognition
Systems analysis
Text processing
Vector quantization
Clean speech
Feature sets
Fusion of classifiers
Identification rates
Principal component classifiers
Principal components
Telephone speech
Text-independent speaker identification
Principal component analysis
Issue Date: 2010
Publisher: IEEE
Citation: Hanilçi, C. ve Ertaş, F. (2010). "Principal component based classification for text-independent speaker identification". ICSCCW 2009 - 5th International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, 39-42.
Abstract: Classification based on Principal Component analysis has recently appeared in the literature in application to text-independent speaker identification. However, results have been reported for only clean speech data. In this paper, we evaluate the performance of principal component classifier for text-independent speaker identification on telephone speech. We then improve its identification performance using a Vector Quantization classifier in combination, through fusion of classifier scores. An identification rate of 78.27% has been obtained on the NTIMIT database, which is well above the best identification rate ever reported in the literature obtained by using only one type of feature set.
Description: Bu çalışma, 02-04 Eylül 2010 tarihleri arasında Famagusta[Kuzey Kıbrıs Türk Cumhuriyeti]’da düzenlenen 5. International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control’da bildiri olarak sunulmuştur.
URI: https://doi.org/10.1109/ICSCCW.2009.5379490
https://ieeexplore.ieee.org/abstract/document/5379490
http://hdl.handle.net/11452/28798
Appears in Collections:Scopus
Web of Science

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.