Animal communication is by nature a dynamic interaction between signaler and audience. The same signal may serve many different functions among different audiences and individuals might tailor signals in different ways depending audience. Singing mice (Scotinomys teguina) are small rodents which have recently captured attention for their loud, long, elaborated song. However, the vocalizations of S. teguina have rarely been studied across social contexts. Here we examine vocalizations of focal S. teguina males produced both alone, in the presence of unfamiliar males, and in the presence of both familiar and unfamiliar females. We find that singing mice produce not one but two distinct types of songs, the well-studied long advertisement song and a shorter, quieter “strophe” song. The songs are produced in different contexts, with strophe songs in particular produced almost entirely in the context of female audiences. Additionally, we find that both advertisement songs and strophe songs change in structure when produced in different circumstances. Finally, we note that both advertisement songs and strophe songs contain considerable information about individual identity, raising the possibility that mice may be able to identify familiar individuals from songs alone.