Traditional active noise reduction (ANC) systems hope to cancel out all the energy of the sound field, ignoring the existence of useful signals. In view of this situation, this paper designs a preprocessing system based on the independent component analysis (ICA) noisy instantaneous mixed model. By using the separation matrix information and the calculation of the output signal kurtosis value, the arrangement ambiguity and scale ambiguity problems are solved, and a pure reference signal is extracted for active noise reduction. Theoretical analysis and simulation results show that the speech signal can still be effectively and accurately separated when the speech signal is less affected by interference and the signal-to-noise ratio of the remaining mixed signals continues to deteriorate.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore