The use of the “Caterpillar” method for the tasks of voice range noise filtering

摘要

This paper, the singular spectral analysis (SSA) method based on the decomposition of the primary time series into additive principal components is used. The algorithm of a mathematical model of the voice selection against the noise background using the SSA method is considered. Graphs of simulation results that characterize the quality of the voice of distorted noise with different intensity are investigated.