Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence
Applied mathematics, Electrical and electronic engineering, Communication engineering and systems
Power-weighted estimators have recently been proposed for relative attenuation and delay estimation in blind source separation. Their provenance lies in the observation that speech is approximately windowed-disjoint orthogonal (WDO) in the time-frequency (TF) domain; it has been reported that using WDO, derived from TF representations of speech, improves mixing parameter estimation. We show that power-weighted relative attenuation and delay estimators can be derived from a particular case of a weighted Bregman divergence. We then propose a wider class of estimators, which we tune to give better parameter estimates for speech.
deFrein, R. & Rickar, S.T. (2016). Power-Weighted Divergences for Relative Attenuation and Delay Estimation. IEEE Signal Processing Letters, vol. 23, no. 11, pg. 1612-1616. doi: 10.1109/LSP.2016.2610481
IRC International Career Development Fellowship co-funded by Marie Curie Actions