Parkinson disease (PD) causes voice and swallow dysfunction even in early stages of the disease. Treatment of this dysfunction is limited, and the neuropathology underlying this dysfunction is poorly defined. Targeted exercise provides the greatest benefit for offsetting voice and swallow dysfunction, and previous data suggest the hypoglossal nucleus and noradrenergic-locus coeruleus (LC) may be involved in its early pathology. To investigate relationships between targeted exercise and neuropathology of voice and swallow dysfunction, we implemented a combined exercise paradigm that included tongue force and vocalization exercises early in the Pink1-/- rat model. We tested the hypotheses that (1) tongue and vocal exercise improves tongue force and timing behaviors and vocalization outcomes, and (2) exercise increases optical density of serotonin (5-HT) in the hypoglossal nucleus, and tyrosine hydroxylase immunoreactive (Th-ir) cell counts in the LC. At two months of age Pink1-/- rats were randomized to exercise or non-exercise treatment. Age-matched wildtype (WT) control rats were assigned to non-exercise treatment. Tongue force and timing behaviors and ultrasonic vocalizations were measured at baseline (two months) and final (four months) timepoints. Optical density of 5-HT in the hypoglossal nucleus and TH-ir cell counts in the LC were obtained. Pink1-/- rats produced greater tongue forces, faster tongue contraction, and higher-intensity vocalization following exercise. There were no differences in LC TH-ir. The non-exercised Pink1-/- group had reduced density of 5-HT in the hypoglossal nucleus compared to the WT control group. The changes to tongue function and vocalization after targeted exercise suggests exercise intervention may be beneficial in early PD.