ANALYSIS OF HARMONIC CONTENTS, FORMANTS, AND VOICE ONSET TIME (VOT) IN THE PRONUNCIATION OF INDONESIAN PHONEMES
Acoustic phonetics in the study of language is related to the biological system of speech production that differentiates timbre. Differences in timbre for each person can be identified through the characteristics of the speech waves. The analysis of timbre in this research is based on physical pa...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/83397 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Acoustic phonetics in the study of language is related to the biological system of speech
production that differentiates timbre. Differences in timbre for each person can be identified
through the characteristics of the speech waves. The analysis of timbre in this research is based
on physical parameters such as fundamental and resonant frequencies as well as the duration
of the phonemes pronunciation. This experiment aimed to characterize the harmonic content
of Indonesian speakers, mapping the vowel [a], [e], [?], [i], [o], and [u] without and with the
influence of consonants, and observing the voice onset time (VOT) of plosive sounds [k] and
[t]. Ten male and female subjects aged 21–23 provided samples of their voices by recording
each vowel phoneme for approximately 2 seconds and the words [sore], [ini], [k?], and [utara]
with a total of 4 iterations. The data is processed using Praat 6.3.07 software. Acoustic signals
and spectrograms are produced, which are then processed to obtain the parameters required.
The harmonic contents are obtained from the sound spectrum, the formants from the
spectrogram, and the sound duration from the acoustic signal and spectrogram. It was found
that the overall harmonic contents for female speakers are in a higher frequency range than for
males due to the shorter length of the vocal tract. The average f1 value for women is 225.5 Hz
and for men 112.2 Hz. The formants of vowel phonemes with the inclusion of consonants
change the data distribution on the vowel chart, especially for the sounds [?], [o], and [u]. The
articulation process can influence that matter. The duration of VOT for [k] is longer than for
[t] for all speakers because the place of articulation is at the velar, near the vocal folds.
|
|---|