Weighted Prediction Error for speech dereverberation.
See ./scripts/sptk/apply_wpe.py -h
echo "egs asset/egs.wav" | ../../scripts/sptk/apply_wpe.py \
--frame-len 512 \
--frame-hop 128 \
- wpe
To use nara-wpe:
echo "egs asset/egs.wav" | ../../scripts/sptk/apply_wpe.py \
--frame-len 512 \
--frame-hop 128 \
--nara-wpe true \
- wpe
To run WPD, using command
echo "egs asset/egs.wav" | ../../scripts/sptk/apply_wpd.py \
--frame-len 512 \
--taps 10 --delay 3 --context 1 \
--wpd-iters 2 --cgmm-iters 10 \
--update-alpha false \
--dump-mask true - wpd
which will generate TF-masks and dereverbrated & enhanced audio simultaneously.
- nara_wpe
- Yoshioka, Takuya, and Tomohiro Nakatani. "Generalization of multi-channel linear prediction methods for blind MIMO impulse response shortening." IEEE Transactions on Audio, Speech, and Language Processing 20.10 (2012): 2707-2720.
- Nakatani, Tomohiro, and Keisuke Kinoshita. "A unified convolutional beamformer for simultaneous denoising and dereverberation." IEEE Signal Processing Letters 26.6 (2019): 903-907.
- Boeddeker, Christoph, et al. "Jointly optimal dereverberation and beamforming." ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2020.
- Nakatani, Tomohiro, and Keisuke Kinoshita. "Maximum likelihood convolutional beamformer for simultaneous denoising and dereverberation." 2019 27th European Signal Processing Conference (EUSIPCO). IEEE, 2019.