BioengineeringbioRxiv
Linear versus deep learning methods for noisy speech separation for EEG-informed attention decoding
N. Das, J. Zegers, H. Van hamme…·A. Bertrand
This paper compares linear and deep learning methods for speech separation in EEG-informed attention decoding, finding linear methods competitive and DNNs beneficial when combined with beamformers.
DOI 10.1088/1741-2552/aba6f8