Production of Mandarin Chinese Tones by Sri Lankan CFL Learners: An Acoustic Analysis

Abstract

Tonal variation in Mandarin forms a relationship with the meanings of words at a lexical level. Tones and tone sandhi are considered distinctive features of Mandarin Chinese phonetics; thus, acquiring accurate Mandarin pronunciation is challenging for speakers of other languages. The present study examined the production of Mandarin tones by Sri Lankan learners through acoustic analysis of f0 using Praat. The study participants were seven (n=7) undergraduate students at intermediate-level Chinese. Each participant recorded a sample of 20 elements, totaling 140 at three tiers: isolation, disyllabic form, and synthesis. The recorded acoustic data were analyzed using Praat 5.4.04 and SPSS Statistics 21. The results indicated that the mean f0 values of the realization of the four tones by the informants are heterogenous at all three tiers. The realization of T3 demonstrated a low accuracy, while T2 was realized with the highest accuracy at all three tiers. As revealed by the tone sandhi analysis, participants have resorted to the underlying representation of the tone sandhi than the surface realization.