Understanding Spoken Language Development of Children with ASD Using Pre-trained Speech Embeddings

TL;DR

This paper explores the use of pre-trained speech embeddings to automatically analyze and classify spoken language development in children with ASD, aiming to improve early assessment methods.

Contribution

It introduces speech processing applications for automated assessment of ASD children's language development, achieving high classification accuracy with pre-trained speech embeddings.

Findings

F1 macro score of 82.6% for child vs. adult speech classification

F1 macro score of 67.8% for speech vs. nonverbal vocalization classification

Demonstrates potential for scalable, automated ASD language assessment tools

Abstract

Speech processing techniques are useful for analyzing speech and language development in children with Autism Spectrum Disorder (ASD), who are often varied and delayed in acquiring these skills. Early identification and intervention are crucial, but traditional assessment methodologies such as caregiver reports are not adequate for the requisite behavioral phenotyping. Natural Language Sample (NLS) analysis has gained attention as a promising complement. Researchers have developed benchmarks for spoken language capabilities in children with ASD, obtainable through the analysis of NLS. This paper proposes applications of speech processing technologies in support of automated assessment of children's spoken language development by classification between child and adult speech and between speech and nonverbal vocalization in NLS, with respective F1 macro scores of 82.6% and 67.8%, underscoring the potential for accurate and scalable tools for ASD research and clinical use.