Continuous Sign Language Recognition and Its Translation into Intonation-Colored Speech

This article is devoted to solving the problem of converting sign language into a consistent text with intonation markup for subsequent voice synthesis phrases by speech intonation. The paper proposes an improved method continuous recognition language, results which are transmitted natural processor based on analyzers morphology, syntax, and semantics Kazakh including morphological inflection construction model simple sentences. approach has significant practical social significance, as it can lead development technologies that will help people disabilities communicate improve their quality life. As result cross-validation model, we obtained average test accuracy 0.97 val_accuracy 0.90 evaluation. We also identified 20 sentence structures intonational model.