EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

KAZMORPHLM: MORPHEME-AWARE LANGUAGE MODEL FOR KAZAKH AUTOMATIC SPEECH RECOGNITION

Yerlan Karabaliyev, Kateryna Kolesnikova

Scientific Journal of Astana IT University · 30.03.2026

DOI: 10.37943/25scdm2312 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific Journal of Astana IT University, online erschienen am 30.03.2026. Verfasst von Yerlan Karabaliyev und 1 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents KazMorphLM, a morpheme-aware language model for automatic speech recognition (ASR) in the Kazakh language. Kazakh belongs to the Turkic family and is characterised by a highly agglutinative morphology, in which a single root can generate a large number of inflected forms through productive suffixation. This property causes severe data sparsity for conventional word-level language models and reduces recognition accuracy. The proposed model introduces three main innovations. First, a rule-based morpheme segmenter uses an inventory of 230 suffixes across fourteen grammatical categories and includes phonological validation through vowel harmony and consonant assimilation rules. Second, a two-level interpolated n-gram architecture combines a 7-gram morpheme-level model with a 5-gram word-level model using an interpolation ratio of 0.6 to 0.4 and Witten–Bell smoothing. Third, a four-channel rescoring mechanism integrates acoustic confidence, word-level and morpheme-level language-model probabilities, and a vowel-harmony consistency score. KazMorphLM was integrated into a hybrid ASR pipeline combining NVIDIA FastConformer and Meta MMS-1B acoustic models. On the FLEURS test set, the system achieves a word error rate of 6.86%, a 14.6% relative improvement over word-level KenLM rescoring. The results indicate that higher-order morpheme modelling is essential for agglutinative languages and that corpus quality outweighs corpus size. The approach is applicable to other morphologically rich Turkic languages.

Zitieren

APA

Karabaliyev, Y., & Kolesnikova, K. (2026). KAZMORPHLM: MORPHEME-AWARE LANGUAGE MODEL FOR KAZAKH AUTOMATIC SPEECH RECOGNITION. Scientific Journal of Astana IT University. https://doi.org/10.37943/25scdm2312

Vancouver

Karabaliyev Y, Kolesnikova K. KAZMORPHLM: MORPHEME-AWARE LANGUAGE MODEL FOR KAZAKH AUTOMATIC SPEECH RECOGNITION. Scientific Journal of Astana IT University. 2026. doi:10.37943/25scdm2312

BibTeX
@article{Karabaliyev2026,
  author = {Karabaliyev, Yerlan and Kolesnikova, Kateryna},
  title = {KAZMORPHLM: MORPHEME-AWARE LANGUAGE MODEL FOR KAZAKH AUTOMATIC SPEECH RECOGNITION},
  journal = {Scientific Journal of Astana IT University},
  year = {2026},
  doi = {10.37943/25scdm2312}
}
Kostenlos starten →