Abstract
The Speech-to-Song Illusion (STS) phenomenon, where repeated short speech utterances transform into perceived song, has drawn attention to its underlying mechanisms and cross-linguistic differences. This study examines the STS effects among Korean speakers, comparing standard Korean (non-tonal) and dialects such as Gyeongsang (pitch-accent, tonal) and Jeju (non-tonal but intonation-rich), which exhibit varying levels of linguistic tonal features. Participants (N = 60), evenly divided between standard and dialect users, evaluated 180 auditory stimuli comprising standard Korean, Gyeongsang, and Jeju utterances under controlled repetition conditions.
Results revealed significant STS effects across all groups and stimuli, with stronger effects observed for dialectal stimuli, particularly Jeju, compared to standard Korean. Interestingly, differences between standard and dialect speaker groups in STS perception were not statistically significant, suggesting that exposure to diverse linguistic environments, facilitated by modern Korean media, may homogenize perceptual responses to tonal variations. The study highlights the influence of tonal and rhythmic elements in STS perception and underscores the cultural and linguistic uniqueness of Korean as a fertile ground for exploring auditory illusions.
This research contributes to understanding the interplay of linguistic and perceptual factors in STS and opens avenues for cross-cultural comparisons and neuroscientific investigations of auditory illusions.