The weighted prediction error (WPE) algorithm has proven to be a very successful dereverberation method for the REVERB challenge. Likewise, neural network based mask estimation for beamforming demonstrated very good noise suppression in the CHiME 3 and CHiME 4 challenges. Recently, it has been shown that this estimator can also be trained to perform dereverberation and denoising jointly. However, up to now a comparison of a neural beamformer and WPE is still missing, so is an investigation into a combination of the two. Therefore, we here provide an extensive evaluation of both and consequently propose variants to integrate deep neural network based beamforming with WPE. For these integrated variants we identify a consistent WER reduction on two distinct databases. In particular, our study shows that deep learning based beamforming benefits from a model-based dereverberation technique (i.e. WPE) and vice versa. Our key findings are: (a) Neural beamforming yields the lower WER in comparison to WPE the more channels and noise are present. (b) Integration of WPE and a neural beamformer consistently outperforms all stand-alone systems.
Cite as: Drude, L., Boeddeker, C., Heymann, J., Haeb-Umbach, R., Kinoshita, K., Delcroix, M., Nakatani, T. (2018) Integrating Neural Network Based Beamforming and Weighted Prediction Error Dereverberation. Proc. Interspeech 2018, 3043-3047, doi: 10.21437/Interspeech.2018-2196
@inproceedings{drude18_interspeech,
author={Lukas Drude and Christoph Boeddeker and Jahn Heymann and Reinhold Haeb-Umbach and Keisuke Kinoshita and Marc Delcroix and Tomohiro Nakatani},
title={{Integrating Neural Network Based Beamforming and Weighted Prediction Error Dereverberation}},
year=2018,
booktitle={Proc. Interspeech 2018},
pages={3043--3047},
doi={10.21437/Interspeech.2018-2196}
}