Speaker identification through use of features selected using genetic algorithm
暂无分享,去创建一个
The authors introduce the use of a genetic algorithm in the reduction of a 24 parameter (12 LPC derived cepstral coefficients + 12 Δ-cepstral coefficients) set to a five, six, seven, eight or ten parameter set, for each speaker in text-independent speaker identification. The experimental results show that there is ~5% increase in the recognition rate when the reduced set of parameters is used.
[1] A. Cohen,et al. On text independent speaker identification using a quadratic classifier with optimal features , 1989, Speech Commun..
[2] Salvatore D. Morgera,et al. Toward a Fundamental Theory of Optimal Feature Selection: Part II-Implementation and Computational Complexit , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.
[3] Ronald W. Schafer,et al. Digital Processing of Speech Signals , 1978 .