Ensemble Learning and Optimizing KNN Method for Speaker Recognition

Ensemble with K Nearest Neighbor (KNN) learner is a novel approach to speaker recognition. It has many advantages over other conversational methods such as simplicity and good generalization ability. At the same time, the generalization ability of an ensemble could be significantly better than that of a single learner. In this paper, we intend to improve the performance of the speaker recognition system by introducing a novel method combining optimizing annular region-weighted distance k nearest neighbor with BagWithProb ensemble learning schemes. Experiments studied in this paper indicate that the proposed method can effectively improve the accuracy of speaker identification system.

[1]  Zhi-HuaZhou,et al.  Adapt Bagging to Nearest Neighbor Classifiers , 2005 .

[2]  Robert Sabourin,et al.  Optimizing nearest neighbour in random subspaces using a multi-objective genetic algorithm , 2004, ICPR 2004.

[3]  Steve Young,et al.  A review of large-vocabulary continuous-speech recognition , 1996 .

[4]  Bayya Yegnanarayana,et al.  Combining evidence from residual phase and MFCC features for speaker recognition , 2006, IEEE Signal Processing Letters.

[5]  Tomi Kinnunen,et al.  Real-time speaker identification and verification , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[6]  Naohiro Ishii,et al.  Text Classification by Combining Different Distance Functions withWeights , 2006, Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD'06).

[7]  Wang Jinfang,et al.  Performance analysis of speaker features extracted from high-order fractional domains , 2005, 2005 International Conference on Industrial Electronics and Control Applications.

[8]  Hui Wang,et al.  Nearest neighbors by neighborhood counting , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Robert Sabourin,et al.  Optimizing nearest neighbour in random subspaces using a multi-objective genetic algorithm , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[10]  G. Miramontes.  Application of an Annular/Sphere Search Algorithm for Speaker Recognition , 2005 .

[11]  King-Ip Lin,et al.  Utilizing indexes for approximate and on-line nearest neighbor queries , 2005, 9th International Database Engineering & Application Symposium (IDEAS'05).

[12]  G. Saha,et al.  An F-Ratio Based Optimization on Noisy Data for Speaker Recognition Application , 2005, 2005 Annual IEEE India Conference - Indicon.

[13]  J. P. Campbell.  Speaker recognition : A tutorial : Automated biometrics , 1997 .

[14]  José Ismael de la Rosa Vargas,et al.  Application of an annular/sphere search algorithm for speaker recognition , 2005, 15th International Conference on Electronics, Communications and Computers (CONIELECOMP'05).

[15]  Lin Lin,et al.  Genetic algorithms and fuzzy approach to Gaussian mixture model for speaker recognition , 2005, 2005 International Conference on Natural Language Processing and Knowledge Engineering.

[16]  Douglas E. Sturim,et al.  Support vector machines using GMM supervectors for speaker verification , 2006, IEEE Signal Processing Letters.

[17]  Nikunj C. Oza,et al.  Online Ensemble Learning , 2000, AAAI/IAAI.

[18]  Steve Young,et al.  A review of large-vocabulary continuous-speech , 1996, IEEE Signal Process. Mag..

[19]  Xin Li,et al.  Speaker identification using speech and lip features , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[20]  M. Faundez-Zanuy,et al.  State-of-the-art in speaker recognition , 2005, IEEE Aerospace and Electronic Systems Magazine.

[21]  Ji-Gang Wang,et al.  An adaptive nearest neighbor algorithm for classification , 2005, 2005 International Conference on Machine Learning and Cybernetics.

[22]  Haibo Hu,et al.  Range Nearest-Neighbor Query , 2006, IEEE Trans. Knowl. Data Eng..