intro algorithm sounds

Adaptive beamforming - algorithm details

Adaptive beamforming techniques are a kind of multi-microphone noise suppression approaches that are based on a linear combination of microphone signals. One of the adaptive beamforming algorithms that is considered in the HearCom project is the Spatially Preprocessed Speech Distortion Weighted Multichannel Wiener Filtering algorithm (SP-SDW-MWF). This algorithm is an extension of the well-known Generalized Sidelobe Canceller and consists of a fixed spatial preprocessor and an adaptive noise cancellation stage. The goal of the adaptive stage is to estimate and subsequently remove the noise components in the speech reference, leading to an enhanced output signal. Thereto, optimal filter weights are computed that minimize a weighted sum of the residual noise energy and a speech distortion measure. The amount of noise suppression versus speech distortion can be controlled by the user through a trade-off parameter in the cost function. The algorithmic variant that is evaluated in the project is a low-cost frequency-domain implementation of a stochastic gradient version of the SP-SDW-MWF algorithm.

References:

Doclo, S., Spriet, A. and Moonen, M. (2004). Efficient frequency-domain implementation of speech distortion weighted multi-channel wiener filtering for noise reduction. In Proc. of the European Signal Processing Conference (EUSIPCO), pages 2007–2010, Vienna, Austria. 

Spriet, A., Moonen, M. and Wouters, J. (2004). Spatially pre-processed speech distortion weighted multi-channel wiener filtering for noise reduction. Signal Processing, 84(12):2367–2387. 

Doclo, S., Spriet, A., Wouters, J. and Moonen, M. (2005). Speech distortion weighted multi-channel wiener filtering techniques for noise reduction . In J. Benesty, S. Makino, J. C., editor, Speech Enhancement (chapter 9), pages 199–228. Springer. 

Spriet, A., Moonen, M. and Wouters, J. (2005). Stochastic gradient based implementation of spatially pre-processed speech distortion weighted multi-channel wiener filtering for noise reduction in hearing aids. IEEE Transactions on Signal Processing, 53(3):911– 925.