Artificial intelligence

Abstract: In the article, features of text- independent approach and the main steps of the speaker identification procedure are considered. A modification of the standard identification scheme using pre-segmentation of the speech signal and the structuring of speaker’s models base, as well as merge a set of acoustic feature classifiers into a single decision rule by boosting algorithm. The proposed modification increases the likelihood of text independent identification and reduce the number of required computations.

Keywords: speaker identification, wide phonetic classes, speaker model, Gaussian mixture models, support vector machines

References:

1. Zhao H. Boosting Image Retrieving Result with GPS Sensor Information / Zhao, Hongwei and Wang, Zhenand Liu, Pingping and Gao, Shan and Nong, Zhimeng // Sensor Letters. – American Scientific Publishers,2014. – V. 12, № 2. – P. 213-217.
2. Sascha E.A. Muenzing. DIRBoost – An algorithm for boosting deformable image registration: Applicationto lung CT intra-subject registration / Sascha E.A. Muenzing, Bram van Ginneken, Max A. Viergever,Josien P.W. Pluim // Medical Image Analysis. – Elsevier, 2014. – V. 18, № 3. – P. 449-459.
3. Pervusin E.A. Obzor osnovnyh metodov raspoznavaniya diktorov / E.A. Pervusin // Matematicheskiestruktury I modelirovanie. − 2011. − № 24. − P. 41-54.
4. Sorokin V.N. Verifikatsiya diktora po spektralno-vremennym parametram rechevogo signala / V.N. Sorokin,A.I. Tsiplihin // Informatsionnye processy. – V. 10, № 2. – P. 87-104.
5. Klymenko M.S. Razrabotka struktury tekstonezavisimoy sistemy identifikatsii diktora // Iskustveniy intelekt.– 2012. – №4. – С. 161–171.
6. Martinez, J. Speaker recognition using Mel frequency Cepstral Coefficients (MFCC) and Vector quantization (VQ)techniques / J. Martinez, H. Perez, E. Escamilla, M.M. Suzuki // Electrical Communications and Computers(CONIELECOMP) 22nd International Conference on 27-29 Feb. 2012. – IEEE, 2012. – P. 248-251.
7. Ali Ali Khalil A.A. Efficient speaker identification from speech transmitted over bluetooth based system /Ali Ali Khalil A.A., Mostafa Saad E.S., El-Nabi M.A., Abd El-Samie F.E. // Computer Engineering &Systems (ICCES), 2013 8th International Conference on 26-28 Nov. 2013. – IEEE, 2013. – P. 190-193.
8. Sadyhov R.H. Modeli gaussovyh smesey dlya verifikatsii diktora po proizvolnoy rechi / R.H. Sadyhov, V.V.Rakush // Doklady BGUIR. – 2003. – №4. – P. 95-103.
9. Bartlett P. Generalization performance of support vector machines and other pattern classifiers / P. Bartlett,J. Shawe-Taylor // Advances in Kernel Methods. MIT Press, 1998. 13 p.
10. Zhu, J. Multi-class adaboost / Zhu, J., Zou, H., Rosset, S., Hastie, T. // Statistics and Its Interface. – 2009. –Vol. 2. – P. 349-360.
11. Electronic resource: http://voxforge.org/ru/downloads