New exciting work from @lilyxuegong , @alex_ander, Fatma Deniz, Keith Johnson, @gallantlab, @FrdricTheuniss1 showing the segmentation of the speech stream occurs mostly at the level of diphones and fMRI can capture dynamics faster than TR rdcu.be/dg9xi@UCBerkeleyNeuro
Just published in JOSS: 'Crowsetta: A Python tool to work with any format for annotating animal vocalizations and bioacoustics data.' doi.org/10.21105/joss.05338