Fine-Tuning Neural ASR Models for Low-Resource Kazakh Childrens Speech with Preprocessing Enhancements

Duisenbekkyzy Z. Rakhimova D. Yessirkepova A. Adali E.
2025 Institute of Electrical and Electronics Engineers Inc.

International Conference on Computer Science and Engineering, UBMK
2025 Issue 2025 405 - 409 pp.

This article presents a comprehensive study of automatic speech recognition (ASR) for Kazakh childrens speech, with a particular focus on the impact of signal preprocessing techniques in a low-resource linguistic environment. Although commercial ASR systems such as Whisper and DeepSpeech have shown recent progress, their accuracy on childrens speech remains limited. We propose a robust ASR pipeline that incorporates Wiener filtering, denoising autoencoders, MFCC/log-Mel features, and transformer-based models (Whisper, Conformer). A novel Kazakh child speech corpus was collected from children aged 2 to 9 via Telegram bots, spontaneous recordings, and curated phrases. We evaluated four neural ASR models under noisy and clean conditions. Our experiments demonstrate that preprocessing significantly improves performance: Whisper achieved WER = 15.4%, CER = 12.3%, Accuracy = 75.2%, BLEU = 0.90. The proposed pipeline is effective for improving ASR accuracy for spontaneous and emotionally variable child speech in low-resource environments.

BLEU , children speech recognition , Conformer , denoising autoencoder , Kazakh ASR , low-resource languages , MFCC , noise reduction , WER , Whisper

Text of the article Перейти на текст статьи

Al Farabi Kazakh National University, Almaty, Kazakhstan
Al Farabi Kazakh National University, Institute of Information and Computational Technologies, Almaty, Kazakhstan
Secondary School 54, Shymkent, Kazakhstan
İstanbul Teknik Üniversitesi, Istanbul, Turkey

Al Farabi Kazakh National University
Al Farabi Kazakh National University
Secondary School 54
İstanbul Teknik Üniversitesi

10 лет помогаем публиковать статьи Международный издатель

Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026