Phonetic Variation in Korean /n/: A Corpus Study*
Received: Feb 05, 2025 ; Revised: Apr 10, 2025 ; Accepted: Apr 17, 2025
Published Online: Apr 30, 2025
ABSTRACT
The purpose of this study is to examine the dynamic acoustic properties of the postconsonantal nasal /n/ in Korean, which may be associated with the phonetic variation arising from three different phonological inputs. The inputs of the three types of /n/ sounds are (i) a canonical as in /kamnamu/ →[kamnamu] “a persimmon tree,” an epenthesized or inserted /kaŋnam_jʌk/ →[kaŋnamnjʌk] “Gangnam Station,” and a nasalized /tɨŋɹok/ → [tɨŋnok] “registration”. In this study, we examine the temporal and spectral properties of these three types of /n/ sounds – both the temporal duration and the first and second formant frequency values – in corpus data. Consonantal duration varies as a function of a particular age group and is characterized by an increase in duration from canonical to inserted /n/. Significant spectral variation is observed within the female group, with the inserted and nasalized types characterized by a low F1 and a high F2, respectively. The results partially support the idea that grammatical knowledge may contribute to phonetic variation, thus suggesting that language users may require more time to process phonologically complex linguistic representations.