Analysis of Vowel Formants and Frequencies for Cognitive Health Assessment: A Preliminary Study

Kevin Kristian (Atma Jaya Ageing Research Center, Atma Jaya Catholic University of Indonesia Center for the Study of Sustainable Community, Atma Jaya Catholic University of Indonesia, Jakarta)
Bryan Santoso (Faculty of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia, Jakarta)
Novelya Indrawan (Faculty of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia, Jakarta)
Nicholas Hardi (Department of Psychiatry and Behavioral Sciences, Faculty of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia, Jakarta)
Djoko Setyanto (Faculty of Bioscience, Technology and Innovation, Atma Jaya Catholic University of Indonesia, Jakarta)
Hadi Sutanto (Faculty of Bioscience, Technology and Innovation, Atma Jaya Catholic University of Indonesia, Jakarta)
Yvonne Suzy Handajani (Atma Jaya Ageing Research Center, Atma Jaya Catholic University of Indonesia, Jakarta)
Yuda Turana (Atma Jaya Ageing Research Center, Atma Jaya Catholic University of Indonesia Department of Neurology, Faculty of Medicine and Health Sciences, Atma Jaya Catholic University of Indonesia, Jakarta)

AKSONA
Vol. 6 No. 2 (2026): JULY 2026

Highlight: Voice parameters did not differ between cognitive impairment groups Combined speech analysis may better assess cognition than voice alone.   ABSTRACT Introduction: Recent studies have shown that, in addition to being cost-effective and non-invasive, voice and speech analysis may play an important role in detecting cognitive impairment, such as dementia. Although numerous studies have investigated this topic, the relationship between voice-related abnormalities and cognitive impairment has not been completely captured. Objective: To investigate the relationship between voice-related abnormalities and cognitive impairment. Methods: A total of 100 participants aged >60 years were selected through purposive sampling. Cognitive function was assessed using the Indonesian version of the Montreal Cognitive Assessment (MoCA-INA), and participants were divided into cognitive impairment and non-cognitive-impairment groups. Acoustic analysis of sustained vowels was conducted using audio analysis software, focusing on formant frequencies, formant centralization ratio, fundamental frequency, and amplitude. Data were analyzed using an independent-samples t-test. Results: The cognitive impairment group demonstrated lower mean values for most formant frequencies. However, no statistically significant differences (p > 0.05) were observed between the groups in formant frequencies, formant centralization ratio, fundamental frequency, or amplitude. Conclusion: No significant association was found between voice-related abnormalities and cognitive impairment. However, subtle variations in voice parameters may still indicate other underlying neurological problems.