DBNET: DOA-driven beamforming network for end-to-end farfield sound source separation
This addresses the problem of far-field sound source separation in reverberant and noisy environments for applications like speech enhancement, but it is incremental as it builds on existing deep learning and acoustic signal processing techniques.
The paper tackled the challenge of end-to-end multi-channel source separation in realistic far-field environments by proposing DBnet, a DOA-driven beamforming network, and its extensions with post masking networks, which outperformed state-of-the-art methods on a challenging dataset.
Many deep learning techniques are available to perform source separation and reduce background noise. However, designing an end-to-end multi-channel source separation method using deep learning and conventional acoustic signal processing techniques still remains challenging. In this paper we propose a direction-of-arrival-driven beamforming network (DBnet) consisting of direction-of-arrival (DOA) estimation and beamforming layers for end-to-end source separation. We propose to train DBnet using loss functions that are solely based on the distances between the separated speech signals and the target speech signals, without a need for the ground-truth DOAs of speakers. To improve the source separation performance, we also propose end-to-end extensions of DBnet which incorporate post masking networks. We evaluate the proposed DBnet and its extensions on a very challenging dataset, targeting realistic far-field sound source separation in reverberant and noisy environments. The experimental results show that the proposed extended DBnet using a convolutional-recurrent post masking network outperforms state-of-the-art source separation methods.