Raqamli Transformatsiya va Sun’iy Intellekt
SO‘ZLOVCHILARNI AJRATISHDA ECAPA-TDNN ASOSIDAGI XUSUSIYATLAR VEKTORLARIDAN FOYDALANISH
1 Muhammad al-Xorazmiy nomidagi Toshkent axborot texnologiyalari universiteti
Zenodo DOI: 10.5281/zenodo.22014404
Keywords
Abstract
So‘zlovchilarni ajratishda xususiyatlar vektorlarini shakllantirish eng muhim bosqichlardan biri hisoblanadi. Zamonaviy chuqur neyron tarmoqlar so‘zlovchining individual xususiyatlarini aniqlash va ularni ifodalashda yuqori aniqlikni ta’minlay oladi. Xususan, x-vector kabi mashhur xususiyatlar vektorlari hozirgi tizimlarda asosiy komponent sifatida qo‘llanib kelmoqda. So‘nggi yillarda TDNN arxitekturasi ustida turli takomillashtirishlar ishlab chiqildi. ECAPA-TDNN modeli esa o‘ziga xos kanal e’tibor mexanizmi va ko‘p qatlamli agregatsiya orqali so‘zlovchini tekshirish vazifalarida yuqori samaradorlikni ko‘rsatdi. Tadqiqot mazkur modelni diarizatsiya sohasiga tatbiq etish imkoniyatlarini namoyish etadi.
References
- S. Gao, M.-M. Cheng, K. Zhao, X. Zhang, M.-H. Yang, and P. H. S. Torr, “Res2Net: A new multi-scale backbone architecture,” IEEE TPAMI, pp. 652–662, 2019.
- J. Thienpondt, B. Desplanques, and K. Demuynck, “Cross-lingual speaker verification with domain-balanced hard prototype mining and language-dependent score normalization,” in Proc. Interspeech, 2020, pp. 756–760.
- B. Desplanques, J. Thienpondt, and K. Demuynck, “ECAPATDNN: Emphasized channel attention, propagation and aggregation in TDNN based speaker verification,” in Proc. Interspeech, 2020, pp. 3830–3834.
- Z. Gao, Y. Song, I. McLoughlin, P. Li, Y. Jiang, and L.-R. Dai, “Improving aggregation and loss function for better embedding learning in end-to-end speaker verification system,” in Proc. Interspeech, 2019, pp. 361–365.
- L. Wan, Q. Wang, A. Papir, and I. Lopez-Moreno, “Generalized end-to-end loss for speaker verification,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 4879–4883, 2018.
- G. Sun, C. Zhang, and P. C. Woodland, “Speaker diarisation using 2d self-attentive combination of embeddings,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5801–5805, 2019.
- E. Hoffer, T. Ben-Nun, I. Hubara, N. Giladi, T. Hoefler, and D. Soudry, “Augment your batch: better training with larger batches,” CoRR, vol. abs/1901.09335, 2019.
- T. J. Park, N. Kanda, D. Dimitriadis, K. J. Han, S. Watanabe, and S. Narayanan, “A review of speaker diarization: Recent advances with deep learning,” 2021, arXiv:2101.09624.
- Xasanov U.K., To‘rayev X.Sh., Tolibova N.D., “So‘zlovchilarni ajratish jarayonlarida nutq signallariga dastlabki ishlov berish”, 153-157b, 2025
- J. Hu, L. Shen, and G. Sun, “Squeeze-and-Excitation networks,” in IEEE/CVF CVPR, 2018, pp. 7132–7141.
- Shukurov K.E., Kaxxarov A.A., Xasanov U.K., “O‘qituvchisiz o‘qitish asosida so‘zlovchilarni ajratish”, “Raqamli transformatsiya: axborot texnologiyalari, sun’iy intellekt va iqtisodiyotda yangi davr”, 442-447b, 2025
- D. S. Park, W. Chan, Y. Zhang, C. Chiu, B. Zoph, E. D. Cubuk, and Q. V. Le, “Specaugment: A simple data augmentation method for automatic speech recognition,” in Proc. Interspeech, 2019, pp. 2613–2617.
- H. Zeinali, K. A. Lee, J. Alam, and L. Burget, “SdSV challenge 2020: Large-scale evaluation of short-duration speaker verification,” in Proc. Interspeech, 2020, pp. 731–735.
- G. Sell, D. Snyder, A. McCree, D. Garcia-Romero, J. Villalba, M. Maciejewski, V. Manohar, N. Dehak, D. Povey, S. Watanabe, and S. Khudanpur, “Diarization is hard: Some experiences and lessons learned for the JHU team in the inaugural DIHARD challenge,” in Proc. Interspeech, 2018, pp. 2808–2812.
- D. Garcia-Romero, G. Sell, and A. McCree, “Magneto: X-vector magnitude estimation network plus offset for improved speaker recognition,” in Proc. Odyssey, 2020, pp. 1–8.
- D. Snyder, D. Garcia-Romero, G. Sell, D. Povey, and S. Khudanpur, “X-vectors: Robust DNN embeddings for speaker recognition,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5329–5333, 2018.
- N. Dawalatabad, S. Madikeri, C. C. Sekhar, and H. A. Murthy, “Novel architectures for unsupervised information bottleneck based speaker diarization of meetings,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 29, pp. 14–27, 2021.
- X. Anguera, S. Bozonnet, N. Evans, C. Fredouille, G. Friedland, and O. Vinyals, “Speaker diarization: A review of recent research,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 20, no. 2, pp. 356–370, 2012.
- N. Ryant, K. Church, C. Cieri, A. Cristia, J. Du, S. Ganapathy, and M. Liberman, “The second DIHARD diarization challenge: Dataset, task, and baselines,” in Proc. Interspeech, 2019, pp. 978– 982.
- J. Deng, J. Guo, N. Xue, and S. Zafeiriou, “ArcFace: Additive angular margin loss for deep face recognition,” in IEEE/CVF CVPR, 2019, pp. 4685–4694.