한국융합학회논문지 (Journal of the Korea Convergence Society)

  • 제13권3호
  • /
  • Pages.43-54
  • /
  • 2022
  • /
  • 2233-4890(pISSN)
  • /
  • 2713-6353(eISSN)
한국융합학회 (Korea Convergence Society)
권호 표지
DOI QR코드

DOI QR Code

실시간 동시통번역의 정책기반 성능 비교 연구

Policy-based performance comparison study of Real-time Simultaneous Translation

  • 이정섭 (고려대학교 컴퓨터학과) ;
  • 문현석 (고려대학교 컴퓨터학과) ;
  • 박찬준 (고려대학교 컴퓨터학과) ;
  • 서재형 (고려대학교 컴퓨터학과) ;
  • 어수경 (고려대학교 컴퓨터학과) ;
  • 이승준 (고려대학교 컴퓨터학과) ;
  • 구선민 (고려대학교 컴퓨터학과) ;
  • 임희석 (고려대학교 컴퓨터학과)
  • Lee, Jungseob (Department of Computer Science and Engineering, Korea University) ;
  • Moon, Hyeonseok (Department of Computer Science and Engineering, Korea University) ;
  • Park, Chanjun (Department of Computer Science and Engineering, Korea University) ;
  • Seo, Jaehyung (Department of Computer Science and Engineering, Korea University) ;
  • Eo, Sugyeong (Department of Computer Science and Engineering, Korea University) ;
  • Lee, Seungjun (Department of Computer Science and Engineering, Korea University) ;
  • Koo, Seonmin (Department of Computer Science and Engineering, Korea University) ;
  • Lim, Heuiseok (Department of Computer Science and Engineering, Korea University)
  • 투고 : 2022.01.10
  • 심사 : 2022.03.20
  • 발행 : 2022.03.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

동시통번역은 문장의 일부만으로 번역을 시작하는 온라인 디코딩으로 지연 대비 번역 성능을 평가 지표로 사용한다. 동시통번역 연구의 공통의 목적은 지연 대비 번역 성능을 높이는 것으로, 지연과 번역 성능 사이의 적절한 절충점을 찾는 것이다. 본 논문은 이러한 동시통번역의 현재 연구 흐름을 반영하여 한국어에서 고정 정책 기반 동시통번역의 비교 실험을 진행하였다. 또한, 한국어에서 동시통번역은 토큰화 과정에서 많은 분절이 발생하여 다른 언어 대비 불필요한 지연이 발생하게 되고, 이를 해결하기 위한 n-gram 토큰화 방안 등의 후속 연구의 필요성에 대해 제시하였다.

Simultaneous translation is online decoding to translates with only subsentence. The goal of simultaneous translation research is to improve translation performance against delay. For this reason, most studies find trade-off performance between delays. We studied the experiments of the fixed policy-based simultaneous translation in Korean. Our experiments suggest that Korean tokenization causes many fragments, resulting in delay compared to other languages. We suggest follow-up studies such as n-gram tokenization to solve the problems.

키워드

과제정보

This work was supported by Institute of Information & communications Technology Planning & Evaluation(IITP) grant funded by the Korea government(MSIT) (No. 2020-0-00368, A Neural-Symbolic Model for Knowledge Acquisition and Inference Techniques) and supported by Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education(NRF-2021R1A6A1A03045425).

참고문헌

  1. Vaswani et al. (2017). Attention is all you need. Advances in neural information processing systems, pp. 5998-6008.
  2. K. Cho & M. Esipova. (2016). Can neural machine translation do simultaneous translation? arXiv preprint arXiv:1606.02012.
  3. Ma et al. (2018) Stacl: Simultaneous translation with implicit anticipation and controllable latency using prefix-to-prefix framework. arXiv preprint arXiv:1810.08398.
  4. B. Zheng, K. Liu, R. Zheng, M. Ma, H. Liu & L. Huang. (2020). Simultaneous translation policies: From fixed to adaptive. arXiv preprint arXiv:2004.13169.
  5. N. Arivazhagan, C. Cherry, W. Macherey & G. Foster. (2020). Retranslation versus streaming for simultaneous translation. arXiv preprint arXiv:2004.03643.
  6. M. Elbayad, L. Besacier & J. Verbeek. (2020). Efficient wait-k models for simultaneous machine translation. arXiv preprint arXiv:2005.08595.
  7. B. Zheng, R. Zheng, M. Ma & L. Huang, "Simpler and faster learning of adaptive policies for simultaneous translation," arXiv preprint arXiv:1909.01559, 2019.
  8. S. Li, J. Hu, B. Wang, X. Shi & Y. Chen. (2021). Xmu's simultaneous translation system at naacl 2021. Proceedings of the Second Workshop on Automatic Simultaneous Translation, pp. 19-23.
  9. C.-C. Chiu & C. Raffel. (2017). Monotonic chunkwise attention," arXiv preprint arXiv:1712.05382.
  10. Arivazhagan et al. (2019). Monotonic infinite lookback attention for simultaneous machine translation. arXiv preprint arXiv:1906.05218.
  11. X. Ma, J. Pino, J. Cross, L. Puzon & J. Gu. (2019). Monotonic multihead attention. arXiv preprint arXiv:1909.12406.
  12. H. Satija & J. Pineau. (2016). Simultaneous machine translation using deep reinforcement learning. ICML 2016 Workshop on Abstraction in Reinforcement Learning.
  13. J. Gu, G. Neubig, K. Cho & V. O. Li. (2016). Learning to translate in real-time with neural machine translation. arXiv preprint arXiv:1610.00388.
  14. B. Zheng, R. Zheng, M. Ma & L. Huang. (2019). Simultaneous translation with flexible policy via restricted imitation learning. arXiv preprint arXiv:1906.01135.
  15. R. J. Weiss, J. Chorowski, N. Jaitly, Y. Wu & Z. Chen. (2017). Sequence-to-sequence models can directly translate foreign speech. arXiv preprint arXiv:1703.08581.
  16. R. Ye, M. Wang & L. Li. (2021). End-to-end speech translation via cross-modal progressive training. arXiv preprint arXiv:2104.10380.
  17. C. Wang, Y. Wu, S. Liu, M. Zhou & Z. Yang. (2020). Curriculum pre-training for end-to-end speech translation. arXiv preprint arXiv:2004.10093.
  18. B. Zhang, I. Titov, B. Haddow & R. (2020). Sennrich. Adaptive feature selection for end-to-end speech translation. arXiv preprint arXiv:2010.08518.
  19. H. Nguyen, F. Bougares, N. Tomashenko, Y. Est'eve & L. Besacier. (2020). Investigating self-supervised pre-training for end-to-end speech translation. Interspeech 2020.
  20. Ren et al. (2020). Simulspeech: End-to-end simultaneous speech to text translation. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp. 3787-3796.
  21. X. Ma, J. Pino & P. Koehn. (2020). Simulmt to simulst: Adapting simultaneous text translation to end-to-end simultaneous speech translation. arXiv preprint arXiv:2011.02048.
  22. A. Karakanta, S. Papi, M. Negri & M. Turchi. (2021). Simultaneous speech translation for live subtitling: from delay to display," arXiv preprint arXiv:2107.08807.
  23. X. Ma, Y.Wang, M. J. Dousti, P. Koehn & J. Pino. (2021). Streaming simultaneous speech translation with augmented memory transformer. ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7523-7527.
  24. D. Liu, M. Du, X. Li, Y. Hu & L. Dai. (2021). The ustc-nelslip systems for simultaneous speech translation task at iwslt 2021. arXiv preprint arXiv:2107.00279.
  25. K. Papineni, S. Roukos, T. Ward & W.-J. Zhu. (2002). Bleu: a method for automatic evaluation of machine translation," Proceedings of the 40th annual meeting of the Association for Computational Linguistics, pp. 311-318.
  26. S. Banerjee & A. Lavie. (2005). Meteor: An automatic metric for mt evaluation with improved correlation with human judgments," Proceedings of the acl workshop on intrinsic and extrinsic evaluation measures for machine translation and/or summarization, pp. 65-72.
  27. C. Cherry & G. Foster. (2019). Thinking slow about latency evaluation for simultaneous machine translation. arXiv preprint arXiv:1906.00048.
  28. C. Raffel, M.-T. Luong, P. J. Liu, R. J. Weiss & D. Eck. (2017). Online and linear-time attention by enforcing monotonic alignments. International Conference on Machine Learning, pp. 2837-2846.
  29. T. Kudo & J. Richardson. (2018). Sentencepiece: A simple and language independent subword tokenizer and detokenizer for neural text processing. arXiv preprint arXiv:1808.06226.
  30. G. Kim. J. Son. J. Kim, H. Lee & H. Lim. (2021). Enhancing Korean Named Entity Recognition With Linguistic Tokenization Strategies. IEEE Access, 9, 151814-151823. https://doi.org/10.1109/ACCESS.2021.3126882
  31. C, Park. S. Eo. H. Moon & H. Lim. (2021, June). Should we find another model?: Improving Neural Machine Translation Performance with ONE-Piece Tokenization Method without Model Modification. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Papers (pp. 97-104).
  32. C. Park. K. Kim, Y. Yang, M. Kang & H. Lim. (2021). Neural spelling correction: translating incorrect sentences to correct sentences for multimedia. Multimedia Tools and Applications, 80(26), 34591-34608. https://doi.org/10.1007/s11042-020-09148-2