International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 183 - Number 12 |
Year of Publication: 2021 |
Authors: Anasuodei Moko, Ledisi G. Kabari |
10.5120/ijca2021921432 |
Anasuodei Moko, Ledisi G. Kabari . Voice Parameter Analysis for Early Detection of Autism in Children. International Journal of Computer Applications. 183, 12 ( Jun 2021), 25-29. DOI=10.5120/ijca2021921432
Autism Spectrum Disorder (ASD) is categorized by social, behavioral and must crucially communication deficiencies. Early detection is necessary for an improved positive result in therapy. This paper proposes a voice parameter analysis for the early detection of ASD in children. The Study used MATLAB programming software to display the signal analysis of an obtained voice data from an autistic and a normal child. The key experimental result shows a major difference in both voices in the time-domain, frequency, and spectrogram signal view. It is seen that the power spectrum (dB) set off with a high pitch level, at -50db for the autistic child, while the normal child is slightly below but closer to -60db and it remained relatively stable for an autistic child along the normalized frequency within 0.58hz -1.0hz. The paper thus recommends early voice screening of Autism Spectrum Disorder (ASD) as a significant first step to identify children at risk of the disease development and will need additional evaluation, intervention, and services.