Calibrate speaker scores on comparable speech durations

A speaker-resemblance cutoff learned from long speech should not be reused on short snippets. The amount of voiced material changes the score and can make an unchanged voice look like a mismatch; extremely short inputs can also fail before scoring. Calibrate query windows with comparable duration and speech coverage against the same reference and channel. Preserve an unassessed result when coverage is inadequate instead of padding silence to satisfy a duration check. This guard reduces unsupported decisions; it does not establish a universal minimum or perceptual accuracy.

Applies when: Reference-voice screening scores utterances or streaming chunks of different lengths. Keywords: speaker similarity duration short speech reference calibration coverage silence