MIMO-DBnet: Multi-channel Input and Multiple Outputs DOA-aware Beamforming Network for Speech Separation
This work addresses speech separation for audio processing applications by eliminating the need for pre-known cues, though it appears incremental as it builds on existing deep learning beamformers with a novel architecture.
The paper tackles the problem of multi-channel speech separation without relying on extra cues like speaker features or directional information, proposing MIMO-DBnet, an end-to-end beamforming network that predicts direction-of-arrival embeddings and beamforming weights from the mixture signal, achieving comprehensive decent improvement over baselines and maintaining performance on high frequency bands despite phase wrapping.
Recently, many deep learning based beamformers have been proposed for multi-channel speech separation. Nevertheless, most of them rely on extra cues known in advance, such as speaker feature, face image or directional information. In this paper, we propose an end-to-end beamforming network for direction guided speech separation given merely the mixture signal, namely MIMO-DBnet. Specifically, we design a multi-channel input and multiple outputs architecture to predict the direction-of-arrival based embeddings and beamforming weights for each source. The precisely estimated directional embedding provides quite effective spatial discrimination guidance for the neural beamformer to offset the effect of phase wrapping, thus allowing more accurate reconstruction of two sources' speech signals. Experiments show that our proposed MIMO-DBnet not only achieves a comprehensive decent improvement compared to baseline systems, but also maintain the performance on high frequency bands when phase wrapping occurs.