Keywords
How to Cite
Abstract
Ushbu maqolada o‘zbek tilida nutqni avtomatik aniqlash (ASR) tizimlarini yaratishda ovozli korpuslarning o‘rni ochiq manbalarda e’lon qilingan ma’lumotlar asosida qiyosiy tahlil etilgan. USC, Common Voice (25.0-versiya), FeruzaSpeech va boshqa resurslar hajmi, spikerlar soni, nutq turi, yozuv standarti hamda foydalanish shartlari bo‘yicha taqqoslangan; korpuslararo umumlashtirish esa e’lon qilingan tajribalar natijalari asosida baholangan. Aniqlanishicha, Common Voice o‘zbek segmentida yozib olingan 265,75 soatning atigi 101,01 soati tasdiqlangan; bitta spikerning 52,09 soatlik yozuvlarida o‘qitilgan model boshqa korpuslarda 76 foizdan yuqori WER (so‘zlar bo‘yicha xatolik darajasi) bergan; o‘quv to‘plamiga USC qo‘shilishi esa usc-test bo‘yicha WER qiymatini kamida 2,5 baravar kamaytirgan. Bosh harf va tinish belgilarini hisobga olish ayrim hollarda WER qiymatini 2 baravardan ortiq o‘zgartirishi ko‘rsatilgan. Korpuslarni tayyorlash va e’lon qilish bo‘yicha amaliy tavsiyalar ishlab chiqilgan.
References
1. Radford A., Kim J. W., Xu T., et al. Robust Speech Recognition via Large-Scale Weak Supervision // Proceedings of the 40th International Conference on Machine Learning (ICML 2023). - 2023. - arXiv:2212.04356.
2. Besacier L., Barnard E., Karpov A., Schultz T. Automatic speech recognition for under-resourced languages: A survey // Speech Communication. - 2014. - Vol. 56. - P. 85–100.
3. Musaev M., Mussakhojayeva S., Khujayorov I., et al. USC: An Open-Source Uzbek Speech Corpus and Initial Speech Recognition Experiments // Speech and Computer: 23rd International Conference, SPECOM 2021. - 2021. - P. 437–447. - DOI: 10.1007/978-3-030-87802-3_40.
4. Povey A., Povey K. FeruzaSpeech: A 60 Hour Uzbek Read Speech Corpus with Punctuation, Casing, and Context // Proceedings of the International Conference on Natural Language and Speech Processing (ICNLSP 2024). - 2024. - P. 360–364. - arXiv:2410.00035.
5. Conneau A., Ma M., Khanuja S., et al. FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech // 2022 IEEE Spoken Language Technology Workshop (SLT). - 2022. - arXiv:2205.12446.
6. Mukhamadiyev A., Khujayarov I., Djuraev O., Cho J. Automatic Speech Recognition Method Based on Deep Learning Approaches for Uzbek Language // Sensors. - 2022. - Vol. 22, No. 10. - Art. 3683. - DOI: 10.3390/s22103683.
7. Vaswani A., Shazeer N., Parmar N., et al. Attention Is All You Need // Advances in Neural Information Processing Systems 30 (NIPS 2017). - 2017. - P. 5998–6008.
8. Gulati A., Qin J., Chiu C.-C., et al. Conformer: Convolution-augmented Transformer for Speech Recognition // Proceedings of Interspeech 2020. - 2020. - P. 5036–5040.
9. Baevski A., Zhou Y., Mohamed A., Auli M. wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations // Advances in Neural Information Processing Systems 33 (NeurIPS 2020). - 2020. - P. 12449–12460.
10. Ko T., Peddinti V., Povey D., Khudanpur S. Audio augmentation for speech recognition // Proceedings of Interspeech 2015. - 2015. - P. 3586–3589.
11. Park D. S., Chan W., Zhang Y., et al. SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition // Proceedings of Interspeech 2019. - 2019. - P. 2613–2617.
12. Ardila R., Branson M., Davis K., et al. Common Voice: A Massively-Multilingual Speech Corpus // Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). - 2020. - P. 4218–4222.
13. Mozilla Data Collective. Common Voice Scripted Speech 23.0 – Uzbek [Elektron resurs]. - 2025. - URL: https://datacollective.mozillafoundation.org/datasets/cmflnuzw7fd2f0vzknodytok4 (murojaat sanasi: 24.10.2026).
14. Mozilla Data Collective. Common Voice Scripted Speech 25.0 – Uzbek (cv-corpus-25.0-2026-03-09) [Elektron resurs]. - 2026. - URL: https://datacollective.mozillafoundation.org/datasets/cmn29fa4y0150o1073fqww73p (murojaat sanasi: 24.10.2026).
15. Mukhamadiyev A., Mukhiddinov M., Khujayarov I., Ochilov M., Cho J. Development of Language Models for Continuous Uzbek Speech Recognition System // Sensors. - 2023. - Vol. 23, No. 3. - Art. 1145. - DOI: 10.3390/s23031145.
16. Khassanov Y., Mussakhojayeva S., Mirzakhmetov A., et al. A Crowdsourced Open-Source Kazakh Speech Corpus and Initial Speech Recognition Baseline // Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021). - 2021. - P. 697–706. - arXiv:2009.10334.
17. gina. Creating Open and Accessible Resources for Uzbek: The Uzbekvoice.ai Project [Elektron resurs] // Mozilla Discourse. - 27.03.2023. - URL: https://discourse.mozilla.org/t/creating-open-and-accessible-resources-for-uzbek-the-uzbekvoice-ai-project/112634 (murojaat sanasi: 26.10.2026).