New preprint: Cloned voices are easier to understand in noise than their human originals: the voice cloning intelligibility benefit (with @hwanguc.bsky.social) osf.io/preprints/ps...

NEW PAPER ALERT! In a dual-task-fMRI paradigm, we used machine learning (XGBoost) as a robust and explainable approach to revealing the dissociable task effects on activation patterns for noise-vocoded speech processing in a set of frontotemporal regions that were hidden by inferential statistics🧠😆.

What a joyful day to wrap up four and a half amazing years at UCL! Huge thanks to my brilliant supervisor @pattiadank.bsky.social, the Speech on the Brain Lab, and everyone who’s been part of the ride🤩 UCL has always been—and will always be—a meaningful part of me😝