This paper aims at (i) characterizing, in a acoustic-phonetic approach, the vocalized /R/ that occurs in medial syllabic coda position, based on data collected with one informant from the countryside cities in São Paulo state and (ii) contrasting occurrences of /R/, the glide [j], the high vowel [i]. It also compares [j] with [i], once some /R/ realizations are similar to [j] and [i] ones. The sample analysis focuses on the frequency of the first three formants (F1, F2 e F3). This data were recorded and we undertook acoustic analyses. The adopted theoretical reference was that of Acoustic Theory of Speech Production by Fant (1960), added to Sociolinguistic framework. The results suggest that (i) before /e/ and /a/, /R/ is vocalized, and (ii) before /ɔ/ and /u/, /R/ is produced as a retroflex sound and does not undergo vocalization.
acoustic phonetics; retroflex; rhotics; vocalization