Differences in formant frequencies between men and women contribute to the perception of voices as masculine or feminine. This study investigated whether visual-acoustic biofeedback can be used to help transgender women achieve formant targets typical of cisgender women, and whether such a shift influences the perceived femininity of speech. Transgender women and a comparison group of cisgender males were trained to produce vowels in a word context while also attempting to make a visual representation of their second formant (F2) line up with a target that was shifted up relative to their baseline F2 (feminized target) or an unshifted or shifted-down target (control conditions). Despite the short-term nature of the training, both groups showed significant differences in F2 frequency in shifted-up, shifted-down, and unshifted conditions. Gender typicality ratings from blinded listeners indicated that higher F2 values were associated with an increase in the perceived femininity of speech. Consistent with previous literature, we found that fundamental frequency and F2 make a joint contribution to the perception of gender. The results suggest that biofeedback might be a useful tool in voice modification therapy for transgender women; however, larger studies and information about generalization will be essential before strong conclusions can be drawn.