Normalized minimum-redundancy and maximum-relevancy based feature selection for speaker verification systems

International Conference
2009-04-19 23:40
Authors : Chi-Sang Jung, Moo-Young Kim, Hong-Goo Kang

Year : 2009

Publisher / Conference : ICASSP

Page : 4549-4552

In this paper, an information theoretical approach to select features for speaker recognition systems is proposed. Conventional approaches having a fixed interval of analysis frames are not appropriate to represent dynamically varying characteristics of speech signals. To maximize the speaker-related information varied by the characteristics of speech signals, we propose an information theory based feature selection method where features are selected to have minimum-redundancy with in selected features but maximumrelevancy to training speaker models. Experimental results verify that the proposed method reduces the error rates of speaker verification systems by 27.37 % in NIST 2002 database.
전체 363
57 International Conference Se-Woon Jeon, Young-cheol Park, Dae Hee Youn "Acoustic depth rendering for 3D multimedia applications" in ICCE, 2012
56 International Conference Jinkyu Lee, Soonho Baek, Hong-Goo Kang "Signal and feature domain enhancement approaches for robust speech recognition" in 8th International Conference on Information, communications and Signal Processing, 2011
55 International Conference Dong-il Hyun, Young-cheol Park, Seok-pil Lee, Dae Hee Youn "Enhanced Interchannel Correlation (ICC) Synthesis for Spatial Audio Coding" in AES 43th International Conference, 2011
54 International Conference Se-Woon Jeon, Young-cheol Park, Seok-Pil Lee, Dae Hee Youn "Virtual Source Panning using Multiple-Wise Vector Base in the Multispeaker Stereo Format" in EUSIPCO, pp.1337-1341, 2011
53 International Conference Dong-il Hyun, Jeongil Seo, Young-cheol Park, Dae Hee Youn "Improved phase parameter analysis and synthesis for parametric stereo audio coding" in ICASSP, 2011
52 International Conference Jeongook Song, Hyen-o Oh, Hong-Goo Kong "Enhanced long-term predictor for Unified Speech and Audio Coding" in ICASSP, 2011
51 International Conference Myung-Suk Song, Cha Zhang, Dinei Florencio, Hong-Goo Kang "Enhancing loudspeaker-based 3D audio with room modeling" in MMSP, 2010
50 International Conference Chi-Sang Jung, Kyu J. Han, Hyunson Seo, Shrikanth S. Narayanan, Hong-Goo Kang "A Variable Frame Length and Rate Algorithm Based on the Spectral Kurtosis Measure for Speaker Verification" in INTERPSEECH, pp.2754-2757, 2010
49 International Conference Ming Li, Chi-Sang Jung, Kyu J. Han "Combining Five Acoustic Level Modeling Methods for Automatic Speaker Age and Gender Recognition" in INTERSPEECH, pp.2826-2829, 2010
48 International Conference Myung-Suk Song , Cha Zhang, Dinei Florencio, Hong-Goo Kang "Personal 3D audio system with loudspeakers" in ICME, 2010