WPD++: An Improved Neural Beamformer For Simultaneous Speech Separation And Dereverberation

Abstract

This paper aims at eliminating the interfering speakers' speech, additive noise, and reverberation from the noisy multi-talker speech mixture that benefits automatic speech recognition (ASR) backend. While the recently proposed Weighted Power minimization Distortionless response (WPD) beamformer can perform separation and dereverberation simultaneously, the noise cancellation component still has the potential to progress. We propose an improved neural WPD beamformer called "WPD++" by an enhanced beamforming module in the conventional WPD and a multi-objective loss function for the joint training. The beamforming module is improved by utilizing the spatio-temporal correlation. A multi-objective loss, including the complex spectra domain scale-invariant signal-to-noise ratio (C-Si-SNR) and the magnitude domain mean square error (Mag-MSE), is properly designed to make multiple constraints on the enhanced speech and the desired power of the dry clean signal. Joint training is conducted to

Stats

• citations7
• S2 citations—
• github stars0
• HF likes0
• heat score6.77
• arxiv keyni2020wpd

Related papers

• A Unified Convolutional Beamformer For Simultaneous Denoising And Dereverberation (2018)14.15
• Joint Multi-channel Dereverberation And Noise Reduction Using A Unified Convolutional Beamformer With Sparse Priors (2021)0.00
• End-to-end Far-field Speech Recognition With Unified Dereverberation And Beamforming (2020)10.61
• Run-time Adaptation Of Neural Beamforming For Robust Speech Dereverberation And Denoising (2024)0.00
• End-to-end Dereverberation, Beamforming, And Speech Recognition With Improved Numerical Stability And Advanced Frontend (2021)10.97
• Sequential Multi-frame Neural Beamforming For Speech Separation And Enhancement (2019)0.00
• Integrating Plug-and-play Data Priors With Weighted Prediction Error For Speech Dereverberation (2023)0.00
• Integrated Speech Enhancement Method Based On Weighted Prediction Error And DNN For Dereverberation And Denoising (2017)0.00