JBT Journal
Advanced

Raqamli Transformatsiya va Sun’iy Intellekt

KVANT DIQQAT MEXANIZMIGA ASOSLANGAN TRANSFORMER MODELIDA MATNNI NUTQQA AYLANTIRISH TIZIMI

Raximov Nodir Odilovich1, Primqulov Oybek Dilmurodovich1, Erkinova Dilnoza Abdurazzoq qizi1

1 Toshkent Axborot Texnalogiyalari Universiteti

Zenodo DOI: 10.5281/zenodo.20065608

Keywords

transformer modeli Matnni nutqqa aylantirish kvant yadro funksiyasi kvant diqqat mexanizmi.

Abstract

Ushbu maqolada kvant diqqat mexanizmiga asoslangan Transformer arxitekturasi yordamida matnni nutqqa aylantirish tizimi taklif etiladi. An’anaviy Transformer modellarida qo‘llaniladigan diqqat mexanizmi murakkab va chiziqli bo’lmagan bog‘lanishlarni ifodalashda imkoniyatlari yetarli darajada emasligi sababli mazkur ishda diqqat mexanizmi kvant yadro funksiyasi bilan takomillashtirildi. Shuningdek, ma’lumotlarni yuqori o‘lchamli Hilbert fazosida qayta ishlash orqali nutq signalining fonetik va prosodik xususiyatlarini kengroq modellashtirishga erishiladi. Taqdim etilgan model o‘zbek tilining o‘ziga xos jihatlari - unli tovushlarning cho‘zilishi, bo‘g‘in ritmi, urg‘u va ohangning muhimligi hamda har bir qo’shimcha aloxida ma’no anglatishi hisobga olgan holda ishlab chiqilgan. Modelda vektorlashtirilgan matn kvant diqqat mexanizmi orqali kontekstual bog‘lanishlar hosil qilinib, davomiylik, chastota va energiya kabi prosodik parametrlar bashorat qilinadi. Yakuniy natijada mel-spektrogramma generatsiya qilinib vocoder yordamida nutq signaliga aylantiriladi. Olingan natijalar kvant diqqat mexanizmi asoslangan Transformer modeli nutq signalining spektral aniqligi, chastota va energiya barqarorligi hamda tabiiyligini ta’minlashda yuqori aniqlikni ta’minlashini isbotladi.

Views: 231 Published: 2026-04-28 Pages: 176-180 Volume: 4 Issue: 2

References

  1. Schuld, M., Sinayskiy, I., Petruccione, F. An Introduction to Quantum Machine Learning // Contemporary Physics, 2015.
  2. Jonathan Shen, Ruoming Pang, Ron J. Weiss, Mike Schuster, Navdeep Jaitly, Zongheng Yang, Zhifeng Chen, Yu Zhang, Yuxuan Wang. Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions // ICASSP, 2018.
  3. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin. Attention Is All You Need // Advances in Neural Information Processing Systems, 2017.
  4. Zhifeng Kong, Wei Ping, Jiaji Huang, Kexin Zhao, Bryan Catanzaro. FastSpeech: Fast, Robust and Controllable Text to Speech // NeurIPS, 2019.
  5. Rabiner, L., Juang, B.H. Fundamentals of Speech Recognition. Prentice Hall, 1993.
  6. Zhifeng Kong, Wei Ping, Jiaji Huang, Kexin Zhao, Bryan Catanzaro. DiffWave: A Versatile Diffusion Model for Audio Synthesis // ICLR, 2021.
  7. Aaron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves. Generative Model for Raw Audio // arXiv preprint arXiv:1609.03499, 2016.
  8. Jacob Biamonte, Peter Wittek, Nicola Pancotti, Patrick Rebentrost, Nathan Wiebe, Seth Lloyd. Quantum Machine Learning // Nature, 2017.
  9. Ho, J., Jain, A., Abbeel, P. Denoising Diffusion Probabilistic Models // NeurIPS, 2020.
  10. Jurafsky, D., Martin, J.H. Speech and Language Processing. Pearson, 2020.
  11. Tokuda, K., Zen, H., Black, A.W. An HMM-based Speech Synthesis System Applied to English // IEEE Speech Synthesis Workshop, 2002.