Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/61489
Title: Speaker recognition algorithm based on channel compensation
Authors: Shen, XJ
Zhai, YJ
Lu, YT
Wang, Y
Chen, HP
Keywords: Computer application
Feature warp
Gaussian mixture model (GMM)
Latent factor analysis (LFA)
Speaker recognition
Support vector machine (SVM)
Issue Date: 2016
Publisher: 吉林工业大学
Source: 吉林大学学报. 工学版, 2016, v. 46, no. 3, p. 870-875 How to cite?
Journal: 吉林大学学报. 工学版 
Abstract: Channel interference factor for the identification results is prevalent among the existing speaker recognition algorithm. In order to improve the accuracy of the system, in this paper, feature warping is used to compensate the channel factor of Mel-Frequency Cepstral Coefficient (MFCC) features. Then, factor analysis technique is applied to deal with the channel factors of the speaker's Gaussian Mixture Model (GMM). In the endpoint detection phase of speech of this recognition system, the GMM for speech modeling is built to accurately determine the beginning and end points of the speech segment, and then the features after feature warping are used to establish speaker GMM. Using factor analysis technique to fit the differences between the speaker characteristics space and the channel space, the algorithm removes channel factor from GMM, and then extracts GMM super-vectors as input of the Support Vector Machine (SVM) to obtain recognition results. Experimental results show that the combination of channel compensation technique and SVM can obtain better recognition rate, and ensure the robustness of the system.
URI: http://hdl.handle.net/10397/61489
ISSN: 1671-5497
DOI: 10.13229/j.cnki.jdxbgxb201603029
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

48
Last Week
1
Last month
Citations as of Oct 21, 2018

Google ScholarTM

Check

Altmetric


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