Evaluation of a feature selection scheme on ICA-based filter-bank for speech recognition

被引:0
作者
Faraji, Neda [1 ]
Ahadi, S. M. [1 ]
机构
[1] Amirkabir Univ Technol, Dept Elect Engn, Tehran, Iran
来源
2007 6TH INTERNATIONAL CONFERENCE ON INFORMATION, COMMUNICATIONS & SIGNAL PROCESSING, VOLS 1-4 | 2007年
关键词
feature extraction; feature selection; filter bank; independent component analysis;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In this paper, we propose a new feature selection scheme that can contribute to an ICA-based feature extraction block for speech recognition. The initial set of speech basis functions obtained in Independent Component Analysis (ICA) training phase, has some redundancies. Thus, finding a minimal-size optimal subset of these basis functions is rather vital. On the contrary to the previous works that used reordering methods on all the frequency bands, we have introduced an algorithm that finds optimal basis functions in each discriminative frequency band. This leads to an appropriate coverage of various frequency components and easy extension to other data is also provided. Our experiments show that the proposed method is very useful, specifically in larger vocabulary size tasks, where the selected basis functions trained using a limited dataset, may get localized in certain frequency bands and not appropriately generalized to residual dataset. The proposed algorithm surmounts this problem by a local reordering method in which contribution of a basis function is specified with three factors: class separability power, energy and central frequency. The experiments on a Persian continuous speech corpus indicated that the proposed method has led to 17% improvement in noisy condition recognition rate in comparison to a conventional MFCC-based system.
引用
收藏
页码:1277 / 1281
页数:5
相关论文
共 7 条
[1]   Feature selection in the independent component subspace for face recognition [J].
Ekenel, HK ;
Sankur, B .
PATTERN RECOGNITION LETTERS, 2004, 25 (12) :1377-1388
[2]  
Gharavian D, 2004, 2004 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL I, PROCEEDINGS, P661
[3]  
HIRSCH HG, 2000, P ISCA ITRW ASR2000, P181
[4]  
Hyvärinen A, 2001, INDEPENDENT COMPONENT ANALYSIS: PRINCIPLES AND PRACTICE, P71
[5]  
KOTANI M, 1999, P INT JOINT C NEUR N, V5, P2981
[6]  
Lee JH, 2000, INT CONF ACOUST SPEE, P1631, DOI 10.1109/ICASSP.2000.862023
[7]   Data-driven spectral basis functions for automatic speech recognition [J].
Malayath, N ;
Hermansky, H .
SPEECH COMMUNICATION, 2003, 40 (04) :449-466