Multi-resolution auditory cepstral coefficient and adaptive mask for speech enhancement with deep neural network

Abstract

The performance of the existing speech enhancement algorithms is not ideal in low signal-to-noise ratio (SNR) non-stationary noise environments. In order to resolve this problem, a novel speech enhancement algorithm based on multi-feature and adaptive mask with deep learning is presented in this paper. First, we construct a new feature called multi-resolution auditory cepstral coefficient (MRACC). This feature which is extracted from four cochleagrams of different resolutions can capture the local information and spectrotemporal context and reduce the algorithm complexity. Second, an adaptive mask (AM) which can track noise change for speech enhancement is put forward. The AM can flexibly combine the advantages of an ideal binary mask (IBM) and an ideal ratio mask (IRM) with the change of SNR. Third, a deep neural network (DNN) architecture is used as a nonlinear function to estimate adaptive mask. And the first and second derivatives of MRACC and MRACC are used as the input of the DNN. Finally, the estimated AM is used to weight the noisy speech to achieve enhanced speech. Experimental results show that the proposed algorithm not only further improves speech quality and intelligibility, but also suppresses more noise than the contrast algorithms. In addition, the proposed algorithm has a lower complexity than the contrast algorithms.

A Hybrid Approach for Single Channel Speech Enhancement using Deep Neural Network and Harmonic Regeneration Noise Reduction

“…A similar method of DNN-based mask estimation was done in past studies [19][20][21][22][23]. A study in [19] modified the feature and proposed adaptive masks in the DNN-based mask estimation with four hidden layers and 1024 hidden nodes. As a result, average PESQ and STOI scores of 2.12 and 0.78, respectively, were obtained at -5dB SNR [19].…”

“…A study in [19] modified the feature and proposed adaptive masks in the DNN-based mask estimation with four hidden layers and 1024 hidden nodes. As a result, average PESQ and STOI scores of 2.12 and 0.78, respectively, were obtained at -5dB SNR [19]. Another study in [23] used the features fusion technique for the DNN input, while the phase-aware and magnitude mask were applied as the target mask.…”