Comparison of Acoustic Parameters of Voice and Speech According to Vowel Type and Suicidal Risk in Adolescents

Globally, suicide prevention and understanding suicidal behavior represent significant health challenges. The predictive potential of voice, speech, and language appears as a promising solution to the difficulty in assessment.

Objective: To analyze variations in acoustic parameters in voice and speech based on vowel types according to different levels of suicidal risk among adolescents in a text reading task.

Methodology: Cross-sectional analytical design using nonprobabilistic sampling. Our sample comprised 98 adolescents aged 14 to 19, undergoing voice acoustic assessment, along with suicidal ideation determination through the Okasha Suicidality Scale and Beck Depression Inventory. Acoustic analysis of recordings was conducted using Praat for phonetic research, Python program, Focusrite interface, and microphone to register voice and speech acoustic parameters such as Fundamental Frequency, Jitter, and Formants. Subsequently, data from adolescents with and without suicidal risk were compared.

Results: Significant differences were observed between suicidal and nonsuicidal adolescents in several acoustic aspects, especially in females in fundamental frequency (F0), signal-to-noise ratio (HNRdB), and temporal variability measured by jitter and standard deviation. In men, differences were found in F0 and HNRdB (P < 0.05).

Conclusion: This study demonstrated statistically significant variations in various voice acoustic parameters among adolescents with and without suicidal risk. These findings underscore the potential relevance of voice and speech as markers for suicidal risk.