This work attempts to explore the association between depression and prosody impairment through acoustic techniques of voice analysis. For this purpose, we selected 10 quantitative vocal parameters, obtained instrumentally, and related to the variability and psychomotor control of phonation, which we have considered as prosody indicators. The sample consisted of 23 patients with diagnosis DSM-III-R of major depression and a control group of 20 healthy volunteers. We obtained results in 7 of the 10 acoustic parameters that suggest reduced prosody in depressed subjects. In a second voice test, carried out 2 months after baseline situation, and while patients were under antidepressant treatment, it was not possible to demonstrate that all these prosody indicators change to normality in those 13 subjects who clinically improved.