Ultrasonic vocalization (USVs) is a promising tool to measure behavioral anxiety in rodents as USV recording is noninvasive, behaviorally relevant, ethological, and reproducible. Studies reporting the effects of stress-induced USVs in adult mice remain limited and debated. We investigated the conditions under which mice emit aversive USVs and evaluated the effects of psychiatric drugs on stress-induced USVs. Male C57BL/6J mice were used. USVs during entire stress sessions were recorded according to their frequency. To investigate the effect of psychiatric drugs on USVs, the number of USVs under cold-restraint stress conditions before and after drug administration was compared. Immediately after stress exposure, blood samples were collected and plasma corticosterone levels were measured. The combination of cold and restraint stress conditions significantly increased the USV numbers and plasma corticosterone levels compared with each stress alone. A benzodiazepine anxiolytic (midazolam) and δ-opioid receptor agonist putative anxiolytic (KNT-127) significantly reduced the stress-induced USV number and plasma corticosterone levels; however, a monoaminergic antidepressant (duloxetine) and N-methyl-D-aspartic acid receptor antagonist antidepressant (ketamine) did not reduce the USV numbers. No changes were noted in the USV numbers after repeated exposure to cold-restraint stress on days 1 and 3. The suppressive effect of midazolam on day 3 was comparable to that on day 1. Stress-induced USV may be used as a quantitative measure of anxiety to systematically assess the effects of anxiolytics. Therefore, cold-restraint stress-induced USVs may be used as a novel tool to measure rodent anxiety and as a useful anxiolytic-screening system.