Automatic Recognition of Dyslalia Affecting Pre-Scholars

TL;DR

This paper presents a system for recognizing dyslalia in pre-school children to facilitate personalized speech therapy, integrating interdisciplinary methods and aiming to improve early intervention for speech impairments.

Contribution

It introduces a recognition component for a comprehensive dyslalia therapy system, combining computer science, psychology, and electronics for personalized treatment.

Findings

Recognition accuracy demonstrated in preliminary tests

System architecture supports real-time processing

Potential to improve early speech disorder diagnosis

Abstract

This article describes the recognition part of a system that will be used for personalized therapy of dyslalia affecting pre scholars. Dyslalia is a speech disorder that affect pronunciation of one ore many sounds. The full system targets interdisciplinary research (computer science, psychology, electronics) - having as main objective the development of methods, models, algorithms, System on Chip architectures with regards to the elaboration and implementation of a complete system addressing the therapy of dyslalia affecting pre scholars, in a personalized and user centered manner. The system addresses the number of 10% of children with age between 4 and 7 that, according to the statistics, present different variations of speech impairments. Although these impairments do not create major difficulties concerning common communication, it has been noticed that problems are likely to appear affecting negatively the child's personality as well as his social environment.