Analysis of Methods of Extracting Feature Vectors of Speech Using Wavelets to Solve Problem of Speaker Voice Identification in Complex Noisy Environment

Basic notions of signal transformations (Fourier, wavelet) are considered. Some models of speech identification are studied that are based on these transformations. In the process of the analysis of identification algorithms of the offered models, main drawbacks and advantages of each model are revealed and recommendations are given on their application under specific noisy conditions.