Among the common acoustic differences between the voices of women and men are the spectral characteristics of fricatives. Akin to pitch and formant frequencies, the frequency characteristics of fricatives are lower in men than in women. These fricative characteristics are learned social cues that help to express the talker’s gender. Listeners reliably incorporate these differences into perceptual judgments by categorizing the same sound differently depending on what they perceive the gender of the talker to be; a fricative that is intermediate to “s” and “sh” will be labeled “s” when thought to be spoken by a man, but “sh” if by a woman. However, the acoustic descriptions and the perceptual consequences of these gender-related acoustic differences have been explored almost exclusively at syllable-onset position. Here we use both word-onset and -offset positions in both perception and production tasks. The perceptual accommodation of talker gender is weaker when the fricative is at syllable offset, inviting questions about whether the extent of gender-related changes in acoustics are simply commensurate with salience of word position. We will present both the perception and production data in an attempt to unify both phenomena in relation to each other.