Effect of anchor modality on the test-retest reliability of voice samples production in acoustic analysis
Acoustic analysis is one of the most common methods in clinical voice assessment. There are a few acoustic parameters widely used in acoustic analysis such as fundamental frequency, jitter, shimmer and harmonic-to-noise ratio. The present study aimed to investigate the effect of anchor modality o...
| Main Author: | |
|---|---|
| Format: | Monograph |
| Language: | English |
| Published: |
Universiti Sains Malaysia
2015
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/60400/ http://eprints.usm.my/60400/1/NUR%20ZARIFHA%20BINTI%20ZAINOL%20ALAM%20-%20e.pdf |
| Summary: | Acoustic analysis is one of the most common methods in clinical voice
assessment. There are a few acoustic parameters widely used in acoustic analysis such
as fundamental frequency, jitter, shimmer and harmonic-to-noise ratio. The present
study aimed to investigate the effect of anchor modality on the test-retest reliability of
voice samples production in each acoustic parameter. This comparative cross-sectional
study recruited 48 female participants by convenience sampling. The data collection
involved two sessions of voice sample recording, i.e., without and with anchor
modality. Firstly, the participants underwent data collection without anchor modality
then followed by data collection with anchor modality with interval of at least one
week between both sessions. For the data collection without anchor modality, the
participants need to produce sustained vowel /a/ at comfortable pitch and loudness
while for data collection with anchor modality, the participants need to produce a
sentence as a standard reference i.e., /satu duwa tiga/ as the anchor modality, and
followed by a sustained vowel /a/ derived from the prolongation of the /a/ of the word
/tiga/. Both situations required the participants to sustain the vowel for five seconds in
three trials. These voice samples were acoustically analysed to obtain the acoustic
parameters. The Intraclass Correlation Coefficients (ICCs) values of each acoustic
parameter from the three trials without and with anchor modality were measured and
compare to determine the effects of the anchor modality on the test-retest reliability of
voice samples production. The results showed that ICC values of trials produced with anchor modality is
almost similar to those produced without anchor modality for the fundamental
frequency, jitter percent and harmonic-to-noise ratio. For fundamental frequency and
harmonic-to-noise ratio, both ICC values for voice produced without and with anchor
modality are rated as excellent meanwhile for jitter percent, both are rated as fair to
good. However, for shimmer percent, there was significant increase in the ICC value
with anchor modality compared to without anchor modality. ICC values in shimmer
percent for voice produced without anchor modality is rated as fair to good while with
anchor modality is rated as excellent. These findings showed a possibility of anchor
modality to improve the consistency of voice samples production in combination with
current clinical practice of using three trials. |
|---|