Speech Enhancement Algorithm Based on Hilbert-Huang and Wavelet

摘要

Combined with Hilbert-Huang transformation of the empirical mode decomposition (EMD) and wavelet analysis, we propose a new speech enhancement algorithm. The algorithm firstly with Hilbert-Huang transformation of the empirical mode decomposition to obtain intrinsic mode functions (IMF), and then combining with wavelet transform of the soft threshold de-noising method, in different intrinsic mode functions on the soft threshold time-scale filtering processing, and finally reconstruct the useful signal to achieve speech enhancement. The simulation results show that the proposed method compared to conventional hard threshold de-noising method, which has more significantly output performance, greatly improved the quality of speech enhancement, and robustness.