On The Role Of Spatial, Spectral, And Temporal Processing For Dnn-based Non-linear Multi-channel Speech Enhancement

Abstract

Employing deep neural networks (DNNs) to directly learn filters for multi-channel speech enhancement has potentially two key advantages over a traditional approach combining a linear spatial filter with an independent tempo-spectral post-filter: 1) non-linear spatial filtering allows to overcome potential restrictions originating from a linear processing model and 2) joint processing of spatial and tempo-spectral information allows to exploit interdependencies between different sources of information. A variety of DNN-based non-linear filters have been proposed recently, for which good enhancement performance is reported. However, little is known about the internal mechanisms which turns network architecture design into a game of chance. Therefore, in this paper, we perform experiments to better understand the internal processing of spatial, spectral and temporal information by DNN-based non-linear filters. On the one hand, our experiments in a difficult speech extraction scenario conf

Stats

• citations10
• S2 citations—
• github stars0
• HF likes0
• heat score7.81
• arxiv keytesch2022on

Related papers

• Insights Into Deep Non-linear Filters For Improved Multi-channel Speech Enhancement (2022)13.93
• Consistency-aware Multi-channel Speech Enhancement Using Deep Neural Networks (2020)0.00
• Decoupled Spatial And Temporal Processing For Resource Efficient Multichannel Speech Enhancement (2024)0.00
• Multichannel Speech Enhancement Without Beamforming (2021)9.41
• Multi-modal Hybrid Deep Neural Network For Speech Enhancement (2016)0.00
• How To Leverage Dnn-based Speech Enhancement For Multi-channel Speaker Verification? (2022)0.00
• Deep Neural Network Techniques For Monaural Speech Enhancement: State Of The Art Analysis (2022)0.00
• Spatial-dccrn: Dccrn Equipped With Frame-level Angle Feature And Hybrid Filtering For Multi-channel Speech Enhancement (2022)5.84