Classification of Autistic and Non-Autistic Children's Speech: A Cross-Linguistic Study in Finnish, French, and Slovak

TL;DR

This study investigates the acoustic features of speech in autistic and non-autistic children across Finnish, French, and Slovak, revealing partially shared speech markers and the challenges of cross-linguistic classification.

Contribution

It provides a cross-linguistic analysis of speech markers of autism, highlighting language-specific and language-general acoustic cues, and evaluates the transferability of classification models across languages.

Findings

Finnish speech model achieved highest accuracy (0.84).

Cross-language models showed moderate success, especially for Slovak and Finnish.

Language-specific acoustic markers of autism were identified.

Abstract

We present a cross-linguistic study of speech in autistic and non-autistic children speaking Finnish, French, and Slovak. We combine supervised classification with within-language and cross-corpus transfer experiments to evaluate classification performance within and across languages and to probe which acoustic cues are language-specific versus language-general. Using a large set of acoustic-prosodic features, we implement speaker-level classification benchmarks as an analytical tool rather than to seek state-of-the-art performance. Within-language models, evaluated with speaker-level cross-validation, yielded heterogeneous results. The Finnish model performed best (Accuracy 0.84, F1 0.88), followed by Slovak (Accuracy 0.63, F1 0.68) and French (Accuracy 0.68, F1 0.56). We then tested cross-language generalization. A model trained on all pooled corpora reached an overall Accuracy of 0.61 and F1 0.68. Leave-one-corpus-out experiments, which test transfer to an unseen language, showed moderate success when testing on Slovak (F1 0.70) and Finnish (F1 0.78), but poor transfer to French (F1 0.42). Feature-importance analyses across languages highlighted partially shared, but not fully language-invariant, acoustic markers of autism. These findings suggest that some autism-related speech cues generalize across typologically distinct languages, but robust cross-linguistic classifiers will likely require language-aware modeling and more homogeneous recording conditions.