Local fuzzy PCA based GMM with dimension reduction on speaker identification
- Authors
- Lee, KY
- Issue Date
- Dec-2004
- Publisher
- ELSEVIER SCIENCE BV
- Keywords
- PCA; GMM; fuzzy clustering; speaker identification; dimension reduction
- Citation
- PATTERN RECOGNITION LETTERS, v.25, no.16, pp.1811 - 1817
- Journal Title
- PATTERN RECOGNITION LETTERS
- Volume
- 25
- Number
- 16
- Start Page
- 1811
- End Page
- 1817
- URI
- http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/19939
- DOI
- 10.1016/j.patrec.2004.07.006
- ISSN
- 0167-8655
- Abstract
- To reduce the high dimensionality required for training of feature vectors in speaker identification, we propose an efficient GMM based on local PCA with fuzzy clustering. The proposed method firstly partitions the data space into several disjoint clusters by fuzzy clustering, and then performs PCA using the fuzzy covariance matrix on each cluster. Finally, the GMM for speaker is obtained from the transformed feature vectors with reduced dimension in each cluster. Compared to the conventional GMM with diagonal covariance matrix, the proposed method shows faster result with less storage maintaining same performance. (C) 2004 Elsevier B.V. All rights reserved.
- Files in This Item
-
Go to Link
- Appears in
Collections - College of Information Technology > ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/19939)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.