One long-standing ornithological paradigm holds that song learning in oscine songbirds has a cultural component, whereas suboscine songbirds inherit songs genetically. Recent studies reveal that suboscine song may be more variable and complex than previously realized. Several suboscine species show marked individual variation in their songs—variation that may play a role in individual recognition and neighbor–stranger discrimination—and a few suboscine species show indications of song learning. We investigated individual variation in the vocalizations of a suboscine passerine, the Screaming Piha (Lipaugus vociferans), from recordings of 26 males at four lek sites along the Tambopata River in Peru. Male Screaming Piha songs consist of quiet introductory syllables followed by two explosively loud syllables that sound like an emphatic pee haw. We used three complementary methods to examine variation in song characteristics. Spectrogram cross-correlation revealed significant consistency within individual males and variability among males. Analysis of fine structural characteristics revealed that all measured song features were significantly less variable within individuals than among individuals. Canonical discriminant analysis based on these 13 song features correctly classified 93.2% of songs by individual and 76.4% of songs by lek site. Our results indicate that there is sufficient consistency in song features within males and sufficient variation among males for identification of individuals on the basis of songs and, to a lesser extent, that song features vary with the lek site of the singer. We conclude that Screaming Pihas sing songs that are individually distinctive and bear a lek signature.