AI-enhanced modeling of phonetic interference and cognitive patterns in azerbaijani-german bilingualism

Abstract

The integration of Artificial Intelligence (AI) into cognitive linguistics has redefined the study of bilingual speech production. This research examines the phonological interplay within the "Azerbaijani-German" pair—two typologically distinct linguistic systems. The study specifically focuses on how the rigid law of vowel harmony (synharmonism) in Azerbaijani affects the acquisition of German consonantal structures and vowel quality in young learners. The study utilized the PRAAT software suite to record and analyze acoustic markers, including F1/F2 formants, pitch variance, and temporal duration. To evaluate lexico-syntactic deviations within children’s media discourse, Natural Language Processing (NLP) models were implemented for high-precision data mining. The occurrence of spontaneous self-correction was identified as a primary marker of emerging linguistic awareness and cognitive control.