Comparison of Acoustic and Perceptual Analysis of Voice in Autistic and Normal Children

Abstract

Background: Autism Spectrum Disorder (ASD) is associated with persistent deficits in communication and social reciprocity, with prosodic and vocal atypicalities frequently reported as clinically relevant features. Prior research has shown inconsistent findings across acoustic and perceptual domains, and there remains a need to clarify the contribution of resonance-based acoustic markers and perceptual voice characteristics in children with ASD. Objective: This study aimed to compare acoustic parameters and perceptual voice features between children with ASD and typically developing peers, with a specific focus on formant frequencies and perceptual ratings as potential discriminators of atypical voice. Methods: An analytical cross-sectional design was employed, enrolling 24 children aged 5–13 years, equally divided between ASD and typically developing groups. Data were collected using Speech Analyzer software and the GRBAS scale. Acoustic variables included duration, intensity, fundamental frequency, and formant frequencies (F1–F3). Perceptual ratings assessed grade, roughness, breathiness, asthenia, and strain, as well as supplementary speech features. Statistical analysis was performed using independent-samples t-tests with effect sizes. Results: No significant group differences were observed in duration, intensity, or fundamental frequency. However, F1 (p = 0.040, d = 0.89) and F2 (p = 0.005, d = 1.27) were significantly lower in the ASD group, with F3 trending toward significance (p = 0.062). Perceptual analysis revealed higher strain in ASD children (p = 0.005, d = 0.89). Additional deficits were observed in clarity, naturalness, pitch modulation, and fluency (all p < 0.001, d > 1.2), alongside increased speech pauses (p = 0.009). Conclusion: Children with ASD exhibit preserved global acoustic features but distinct resonance-based and perceptual voice abnormalities, particularly reduced formant frequencies and elevated strain, which may serve as sensitive clinical markers. Integration of acoustic–perceptual assessments in healthcare can enhance early detection and guide tailored interventions to improve communication outcomes