Using Artificial Intelligence Algorithms for Speech Therapy Systems

Abstract

This article analyzes the algorithms used to process speech signals. A structure for processing Uzbek speech signals has been proposed. Based on this structure, the effectiveness of recognizing Uzbek speech sounds, syllables and words was evaluated on the basis of intellectual algorithms. Based on the evaluated indicators, the structure of neural networks that recognize children's speech pronunciation was developed. Recognition indicators of children’s speech sounds, syllables, and words were identified.