论文信息 - Research on speaker feature dimension reduction based on CCA and PCA

Research on speaker feature dimension reduction based on CCA and PCA

A method to reduce feature dimension based on CCA and PCA is proposed. First, using the CCA to fuse the LPC features based on channel model and the MFCC feature based on auditory model to improve the relevance of the two different features; second, utilizing the PCA to further remove redundant features, and reduce the dimension of effective features. To verify the validity of this method, experimental model is based on GMM speaker recognition system, and 16-dimensional LPC and 13-dimensional MFCC are selected as speaker features. Compared with the traditional dimension reduction method, such as CCA, PCA and manual methods, experiments show that CCA+PCA method can further enhance the effect of dimension reduction.

Xiongwei Zhang | Yuhuan Zhou | Yong Gong | Jinming Wang

[1] Min-Seok Kim,et al. Robust Text-Independent Speaker Identification Using Hybrid PCA&LDA , 2006, MICAI.

[2] Chengjun Liu,et al. Gabor-based kernel PCA with fractional power polynomial models for face recognition , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3] Sun Quan,et al. The Theory of Canonical Correlation Analysis and Its Application to Feature Fusion , 2005 .

[4] Zhang Xiong-wei. Study and Simulation of the Acoustic Features in Speaker Recognition System , 2003 .

[5] Yang Da. Study of Feature Selection for Speech Recognition , 2003 .

[6] Engin Avci,et al. A new optimum feature extraction and classification method for speaker recognition: GWPNN , 2007, Expert Syst. Appl..

[7] Hong Quan. Sub-pattern Canonical Correlation Analysis with Application in Face Recognition , 2008 .

[8] Roberto Togneri,et al. Robust speaker identification using combined feature selection and missing data recognition , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.