论文信息 - Speaker Recognition Method Based on CPSO Clustering and KMP Algorithm

Speaker Recognition Method Based on CPSO Clustering and KMP Algorithm

This paper proposes a chaotic particle swarm optimization and kernel matching pursuit algorithm a combination of speaker recognition methods. First through the CPSO clustering algorithm to transform the MFCC feature parameters of processing, streamlining of the MFCC feature parameters (SMFCC), then use the KMP algorithm on the reduction parameters after the SMFCC feature classification training and recognition. The results show that CPSO-KMP-based speaker recognition method compared to the GMM-UBM method, the relative increase in EER performance of 31%.

Huaitao Shi | Zhe Yuan | Dong An | Meng Shao | Qingchen Pan

[1] Yang Zhang,et al. A novel pedestrian detection method based on Cost-Sensitive Support Vector Machine and Chaotic Particle Swarm Optimization with T mutation , 2012 .

[2] Tomi Kinnunen,et al. Comparison of clustering methods: A case study of text-independent speaker modeling , 2011, Pattern Recognit. Lett..

[3] Grzegorz Dudek. Neuro-fuzzy approach to the next day load curve forecasting , 2011 .

[4] Riccardo Poli,et al. Particle swarm optimization , 1995, Swarm Intelligence.

[5] Seiichi Nakagawa,et al. PAPER Special Section on Processing Natural Speech Variability for Improved Verbal Human-Computer Interaction Speaker Recognition by Combining MFCC and Phase Information in Noisy Conditions , 2010 .

[6] Francis R. Bach,et al. Feature Space Mahalanobis Sequence Kernels: Application to SVM Speaker Verification , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[7] Pascal Vincent,et al. Kernel Matching Pursuit , 2002, Machine Learning.