Zgadzam się
Nasza strona zapisuje niewielkie pliki tekstowe, nazywane ciasteczkami (ang. cookies) na Twoim urządzeniu w celu lepszego dostosowania treści oraz dla celów statystycznych. Możesz wyłączyć możliwość ich zapisu, zmieniając ustawienia Twojej przeglądarki. Korzystanie z naszej strony bez zmiany ustawień oznacza zgodę na przechowywanie cookies w Twoim urządzeniu.
This paper provides a comprehensive review of vowel harmony (VH) in the Kazakh lan‐
guage from a computational linguistics perspective. As an agglutinative member of the Turkic
language family, Kazakh exhibits a highly developed morphological structure and a complex
phonological system. Among its key features, vowel harmony – where suffix vowels change
between back and front depending on the root vowel – is one of the core phonetic laws that
ensures naturalness and phonological coherence. However, building computational models
that accurately capture and automate this phenomenon remains a significant challenge. The
two main types of harmony – palatal (back‐front) and labial (rounding) – interact in complex
ways, and the gradient (non‐binary) nature of labial harmony poses difficulties for traditional
rule‐based and finite‐state models. This review discusses modern statistical and cognitive‐
phonological approaches, particularly the Maximum Entropy Harmonic Grammar (MaxEnt HG)
and Dispersion Theory, which can effectively capture probabilistic and gradient aspects of vowel
harmony. Furthermore, it explores the potential of neural and hybrid modelling frameworks
to capture the phonological patterns of Kazakh with greater empirical precision. Such ap‐
proaches may contribute to improving phonologically aware NLP systems for morphologically
rich and low‐resource languages like Kazakh.
