OPPO’s Machine Translation System for the IWSLT 2020 Open Domain Translation Task

Proceedings of the 17th International Conference on Spoken Language Translation (IWSLT), pages 114–121 July 9-10, 2020. c⃝2020 Association for Computational Linguistics https://doi.org/10.18653/v1/P17 114 OPPO’s Machine Translation System for the IWSLT 2020 Open Domain Translation Task Qian Zhang and Tingxun Shi and Xiaopu Li and Dawei Dang and Di Ai and Zhengshan Xue and Jie Hao {zhangqian666, shitingxun, lixiaopu, dangdawei aidi1, xuezhengshan, haojie}@oppo.com Manifold Lab, OPPO Research Institute Abstract In this paper, we demonstrate our machine translation system applied for the Chinese­ Japanese bi­directional translation task (aka. open domain translation task) for the IWSLT 2020 (Ansari et al., 2020). Our model is based on Transformer (Vaswani et al., 2017), with the help of many popular, widely­proved effective data preprocessing and augmentation methods. Experiments show that these methods can im­ prove the baseline model steadily and signifi­ cantly. 1 Introduction Machine translation, proposed even before the first computer was invented (Hutchins, 2007), has been always a famous research topic of computer sci­ ence. In the recently years, with the renaissance of neural network and the emergence of atten­ tion mechanism (Sutskever et al., 2014) (Bahdanau et al., 2014), the old area has stepped into a new era. Furthermore, the Transformer architecture, after being published, has immediately attracted much attention nowadays and is dominating the whole field now. Although Transformer has achieved many SOTA results, it has tremendous amount of param­ eters so is hard to be fit on small datasets, there­ fore it has a high demand on good and large data source. Despite of the lack of high quality data of parallel corpus, the document level comparable data is relatively easy to be crawled, so exploring an effective and accurate way of mining aligned sentence pairs from such large but noisy data, to enrich the small parallel corpus, could benefit the machine translation system a lot. Besides, cur­ rently, most open access big volume machine trans­ lation datasets are based on English, and many of them are translated to/from another European lan­ guage – As many popular European languages are fusional languages, most corpus are composed by two fusional languages together. To understand whether the existing model architectures and train­ ing skills can be applied on the translation between Asian languages and other type of languages , such as between an analytic language like Chinese and an agglutinative language like Japanese, interest us. In this paper, we demonstrate our system ap­ plied for the IWSLT 2020 open domain text transla­ tion task, which aims to translate Chinese from/to Japanese 1. Besides describing how we trained the model that is used to generate the final result, this paper also introduces how do we mine extra paral­ lel sentences from a large but noisy data released by the organizer, and several experiments inspired by the writing systems of Chinese and Japanese. 2 Data Preprocessing Four pairs of parallel data are provided in the cam­ paign, which are • The very original file, which is crawled from various websites, and is very huge. Ac­ …

Content truncated to 500 words