Neural Network-augmented Kalman Filtering For Robust Online Speech Dereverberation In Noisy Reverberant Environments

Abstract

In this paper, a neural network-augmented algorithm for noise-robust online dereverberation with a Kalman filtering variant of the weighted prediction error (WPE) method is proposed. The filter stochastic variations are predicted by a deep neural network (DNN) trained end-to-end using the filter residual error and signal characteristics. The presented framework allows for robust dereverberation on a single-channel noisy reverberant dataset similar to WHAMR!. The Kalman filtering WPE introduces distortions in the enhanced signal when predicting the filter variations from the residual error only, if the target speech power spectral density is not perfectly known and the observation is noisy. The proposed approach avoids these distortions by correcting the filter variations estimation in a data-driven way, increasing the robustness of the method to noisy scenarios. Furthermore, it yields a strong dereverberation and denoising performance compared to a DNN-supported recursive least squares

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylemercier2022neural

Related papers

• Integrated Speech Enhancement Method Based On Weighted Prediction Error And DNN For Dereverberation And Denoising (2017)0.00
• Run-time Adaptation Of Neural Beamforming For Robust Speech Dereverberation And Denoising (2024)0.00
• Convolutive Prediction For Monaural Speech Dereverberation And Noisy-reverberant Speaker Separation (2021)11.39
• VACE-WPE: Virtual Acoustic Channel Expansion Based On Neural Networks For Weighted Prediction Error-based Speech Dereverberation (2021)3.58
• Neural Kalman Filtering For Speech Enhancement (2020)2.26
• Integrating Plug-and-play Data Priors With Weighted Prediction Error For Speech Dereverberation (2023)0.00
• Deep Speech Enhancement For Reverberated And Noisy Signals Using Wide Residual Networks (2019)0.00
• A Wavenet For Speech Denoising (2017)18.47