Abstract

It is well known that vowels can be produced in isolation, acoustically stable in such a way that they are represented as points in the F1‐F2‐F3 space. Vietnamese language presents 13 vowels, however, Vietnamese can only pronounce 9 vowels [i, u, e, o, ω, a, σ, ε, γ] in isolated mode. A previous study showed that the 4 remaining vowels [α, δ , ν , ε ] have the same acoustic characteristics (F1, F2, F3) as, respectively, the vowels [a, ε, σ, γ], but their dynamic characteristics (the rate of CV transitions) are clearly distinct. Measurements show us that vowels durations of [α, ν] are always shorter than the one of corresponding classical vowels [a, γ] and are not acoustically stable. For test perceptions (with 10 Vietnamese people), synthesized syllables [a‐t] and [γ‐t] with changing vowel duration are recognized as [α‐t, ν‐t] when then duration of initial vowel [a, γ] are 60% ‐ 70% shorter. It means that the vowel duration is an important parameter that allows Vietnamese distinguishing the long vowels and short vowels in Vietnamese language. This paper further analyses the production of Vietnamese CV, including long vowels and short vowels in terms of duration, formant evolution and rate of C.V transitions.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call