DESNet: A Multi-channel Network for Simultaneous Speech Dereverberation, Enhancement and Separation
This work addresses speech processing challenges for applications like hearing aids or communication systems, but it is incremental as it builds on existing methods like E2E-UFE and DCCRN.
The paper tackles the problem of simultaneous speech dereverberation, enhancement, and separation by proposing DESNet, a multi-channel network that outperforms DCCRN and most state-of-the-art structures in non-dereverberated cases and shows improvements over cascaded WPE-DCCRN networks in dereverberated scenarios.
In this paper, we propose a multi-channel network for simultaneous speech dereverberation, enhancement and separation (DESNet). To enable gradient propagation and joint optimization, we adopt the attentional selection mechanism of the multi-channel features, which is originally proposed in end-to-end unmixing, fixed-beamforming and extraction (E2E-UFE) structure. Furthermore, the novel deep complex convolutional recurrent network (DCCRN) is used as the structure of the speech unmixing and the neural network based weighted prediction error (WPE) is cascaded beforehand for speech dereverberation. We also introduce the staged SNR strategy and symphonic loss for the training of the network to further improve the final performance. Experiments show that in non-dereverberated case, the proposed DESNet outperforms DCCRN and most state-of-the-art structures in speech enhancement and separation, while in dereverberated scenario, DESNet also shows improvements over the cascaded WPE-DCCRN networks.