Skip to main content
SpringerLink
Log in

Multilingual sequence to sequence convolutional machine translation

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

In this paper, to improve the translation quality of a sentence, Convolutional sequence to sequence architecture has been applied to English-Punjabi, Punjabi-English, Hindi-Punjabi, Punjabi-Hindi language pairs. The Convolution architecture consists of Gated Linear Unit (GLU) and a Multi-Hop attention mechanism. The GLU mechanism controls the information flow through hidden units to produce good translations. The Multi-Hop attention is an enhanced version of the attention mechanism that allows the model to make repeated gaze on the sentence than looking at the sentence only once. As compared to the Recurrent Neural Network (RNN), the Convolution based technique provides better accuracy, captures long-range dependencies between the words, capture local context, and allows parallelization over every element in a sequence without depending on the computations of the previous time steps. In this work, we have applied the architecture on four Indian language pairs and compared the performance of our model with conventional models using BLEU, METEOR and WER as evaluation metrics. Our model outperforms recurrent neural networks by 2 BLEU points on English-Punjabi, 2.5 BLEU points on Punjabi-English, 3.5 BLEU points on Hindi-Punjabi translations, and 1.5 BLEU points on Punjabi-Hindi.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19
Fig. 20

Similar content being viewed by others

Notes

  1. https://tdil-dc.in/index.php?lang=en

  2. https://github.com/anoopkunchukuttan/indic_nlp_library

  3. https://github.com/facebookresearch/fairseq

  4. https://github.com/moses-smt/mosesdecoder/blob/master/scripts/generic/multi-bleu.perl

  5. https://github.com/zszyellow/WER-in-python

References

  1. Agrawal R (2017) Towards efficient Neural Machine Translation for Indian Languages. Diss. International Institute of Information Technology, Hyderabad

  2. Chen SF, Beeferman D, Rosenfeld R (1998) Evaluation metrics for language models. DARPA Broadcast News Transcription and Understanding Workshop

  3. Dauphin YN, Fan A, Auli M, Grangier D (2017) Language modeling with gated convolutional networks. In International Conference on Machine Learning pp. 933–941

  4. Forcada ML et al (2011) Apertium: a free/open-source platform for rule-based machine translation. Mach Transl 25.2:127–144

    Article  Google Scholar 

  5. Gehring J et al (2017) Convolutional sequence to sequence learning. Proceedings of the 34th International Conference on Machine Learning-Volume 70. JMLR. org

  6. Goyal V, Lehal GS (2011) Hindi to Punjabi machine translation system. Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies: Systems Demonstrations. Association for Computational Linguistics

  7. Goyal V, Lehal GS (2019) Hindi-Punjabi Machine Transliteration System (For Machine Translation System). George Ronchi Foundation Journal, Italy 64.1:2009

    Google Scholar 

  8. He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition pp. 770–778

  9. Jindal S, Goyal V, Bhullar JS (2018) English to Punjabi statistical machine translation using moses (Corpus Based). J Stat Manag Syst 21.4:553–560

    Google Scholar 

  10. Josan GS, Lehal GS (2008) A Punjabi to Hindi machine translation system. 22nd International Conference on on Computational Linguistics: Demonstration Papers. Association for Computational Linguistics

  11. Josan GS, Lehal GS (2012) Direct Approach for Machine Translation from Punjabi to Hindi. CSI J Comput 1.1:3–19

    Google Scholar 

  12. Josan GS, Kaur J (2011) Punjabi to Hindi statistical machine transliteration. Int J Inf Technol Knowl Manag 4.2:459–463

    Google Scholar 

  13. Kaur K (2011) A Punjabi to English machine translation system for legal documents

  14. Kaur J, Josan G (2011) Statistical approach to transliteration from english to punjabi. Int J Comput Sci Eng 3.4:1518–1527

    Google Scholar 

  15. Kaur D, Kaur ER (2015) English To Punjabi Script Converter System For Proper Nouns Using Hybrid Approach. International Journal of Scientific Research and Management 3.3

  16. Kumar A, Goyal V (2018) Hindi to Punjabi machine translation system based on statistical approach. J Stat Manag Syst 21(4):547–552

    Google Scholar 

  17. Kaur A, Rani J (2015) A web based Punjabi to Hindi Statistical Machine Translation System. 2015 2nd International Conference on Recent Advances in Engineering & Computational Sciences (RAECS). IEEE

  18. Koehn P (2009) Statistical machine translation. Cambridge University Press

  19. Koehn P, Hoang H, Birch A, Callison-Burch C, Federico M, Bertoldi N, Herbst E (2007) Moses: Open source toolkit for statistical machine translation. In Proceedings of the 45th annual meeting of the ACL on interactive poster and demonstration sessions 177–180

  20. Kumar P, Kumar V (2013) Statistical machine translation based punjabi to english transliteration system for proper nouns. Int J Appl Innov Eng Manag 2(8):318–321

    Google Scholar 

  21. Lavie A, Agarwal A (2007) METEOR: An automatic metric for MT evaluation with high levels of correlation with human judgements. Proceddings of the Second Workshop on Statistical Machine Translation. Association for Computational Linguistics 228–231

  22. Luong MT, Pham H, Manning CD (2015) Effective approaches to attention-based neural machine translation. arXiv preprintarXiv:1508.04025

  23. Mahata SKU, Das D, Bandyopadhyay S (2019) MTIL2017: Machine translation using recurrent neural network on statistical machine translation. J Intell Syst 28.3:447–453

    Article  Google Scholar 

  24. Nair LR, David Peter S (2012) Machine translation systems for Indian languages. Int J Comp Appl 39.1:0975–8887

    Google Scholar 

  25. Papineni K, Roukos S, Ward T, Zhu W (2002) BLEU: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting on association for computational linguistics, pp.311–318

  26. Premjith B, Kumar MA, Soman KP (2019) Neural Machine Translation System for English to Indian Language Translation Using MTIL Parallel Corpus: Special Issue on Natural Language Processing. J Intell Syst 28(3):387–398

  27. Revanuru K, Turlapaty K, Rao S (2017) Neural Machine Translation of Indian Languages. Proceedings of the 10th Annual ACM India Compute Conference on ZZZ. ACM

  28. Shirai S, Bond F, Takahashi Y (1997) A hybrid rule and example-based method for machine translation. Recent Advances in Example-Based Machine Translation 211–224

  29. Sidhu BK, Singh A, Goyal V (2010) Identification of proverbs in Hindi text corpus and their translation into Punjabi. Journal of Computer Science and Engineering 2(1):32–37

    Google Scholar 

  30. Singh S, Anand Kumar M, Soman KP (2018) Attention based English to Punjabi neural machine translation. J Intell Fuzzy Syst 34.3:1551–1559

    Article  Google Scholar 

  31. Somers H (1999) Example-based machine translation. Mach Transl 14(2):113–157

    Article  MathSciNet  Google Scholar 

  32. Sutskever I, Vinyals O, Le QV (2014) Sequence to sequence learning with neural networks. Advances in neural information processing systems

  33. van der Wel E, Ullrich K (2017) Optical music recognition with convolutional sequence-to-sequence models. arXiv preprintarXiv:1707.04877

  34. Zhang Y et al (2019) Abstract Text Summarization with a Convolutional Seq2seq Model. Appl Sci 9.8:1665

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by Council of Scientific and Indian Research (CSIR) with file no. 09/263(1152)/2018-EMR-I. We are also thankful to Department of Science and Technology (DST), Govt. of India for support through PURSE Grant.

Author information

Authors and Affiliations

  1. School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, India

    Mani Bansal & D. K. Lobiyal

Authors
  1. Mani Bansal
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. D. K. Lobiyal
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mani Bansal.

Ethics declarations

Conflict of interest

Authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bansal, M., Lobiyal, D.K. Multilingual sequence to sequence convolutional machine translation. Multimed Tools Appl 80, 33701–33726 (2021). https://doi.org/10.1007/s11042-021-11345-6

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-021-11345-6

Keywords

Search

Navigation