# Neural Network-augmented Kalman Filtering for Robust Online Speech Dereverberation in Noisy Reverberant Environments

arXiv: https://arxiv.org/abs/2204.02741 · HTML: https://mlsanity.com/paper/2204.02741/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 35 references that are in the map.
- First posted on arXiv 6 Apr 2022.

## Builds on (most cited first, 7 of 7)

- [WHAM!: Extending Speech Separation to Noisy Environments](https://mlsanity.com/paper/1907.01160/) (2019)
- [WHAMR!: Noisy and Reverberant Single-Channel Speech Separation](https://mlsanity.com/paper/1910.10279/) (2019)
- [Coherent-to-Diffuse Power Ratio Estimation for Dereverberation](https://mlsanity.com/paper/1502.03784/) (2015)
- [A unified convolutional beamformer for simultaneous denoising and dereverberation](https://mlsanity.com/paper/1812.08400/) (2018)
- [Low complexity online convolutional beamforming](https://mlsanity.com/paper/2107.06775/) (2021)
- [Integrated sidelobe cancellation and linear prediction Kalman filter for joint multi-microphone speech dereverberation, interfering speech cancellation, and noise reduction](https://mlsanity.com/paper/1906.07512/) (2019)
- [End-To-End Deep Learning-Based Adaptation Control for Frequency-Domain Adaptive System Identification](https://mlsanity.com/paper/2106.01262/) (2021)

## Built on by (most cited first, 1 of 1)

- [Extending DNN-based Multiplicative Masking to Deep Subband Filtering for Improved Dereverberation](https://mlsanity.com/paper/2303.00529/) (2023)

## Abstract (arXiv)

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 variant of WPE, especially for highly noisy inputs.