English-to-Chinese Transliteration with Phonetic Auxiliary Task

Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, pages 378–388 December 4 - 7, 2020. c⃝2020 Association for Computational Linguistics 378 English-to-Chinese Transliteration with a Phonetic Auxiliary Task Yuan He∗ Department of Computer Science University of Oxford yuan.he@cs.ox.ac.uk Shay B. Cohen School of Informatics University of Edinburgh scohen@inf.ed.ac.uk Abstract Approaching named entities transliteration as a Neural Machine Translation (NMT) prob- lem is common practice. While many have applied various NMT techniques to enhance machine transliteration models, few focus on the linguistic features particular to the relevant languages. In this paper, we investigate the effect of incorporating phonetic features for English-to-Chinese transliteration under the multi-task learning (MTL) setting—where we define a phonetic auxiliary task aimed to im- prove the generalization performance of the main transliteration task. In addition to our system, we also release a new English-to- Chinese dataset and propose a novel evalua- tion metric which considers multiple possible transliterations given a source name. Our re- sults show that the multi-task model achieves similar performance as the previous state of the art with a model of a much smaller size.1 1 Introduction Transliteration, the act of mapping a name from the orthographic system of one language to another, is directed by the pronunciation in the source and target languages, and often by historical reasons or conventions. It plays an important role in tasks like information retrieval and machine translation (Mar- ton and Zitouni, 2014; Hermjakob et al., 2008). Over the recent years, many have ad- dressed transliteration using sequence-to-sequence (seq2seq) deep learning models (Rosca and Breuel, 2016; Merhav and Ash, 2018; Grundkiewicz and Heafield, 2018), enhanced with several NMT tech- niques (Grundkiewicz and Heafield, 2018). How- ever, this recent work neglects the most crucial feature for transliteration, i.e. pronunciation. To *Work done at The University of Edinburgh. 1Our code and data are available at https://github. com/Lawhy/Multi-task-NMTransliteration. English IPA Chinese Pinyin A /"eI./ 艾 `ai my /mi/ 米 mˇı Table 1: An example of English-to-Chinese transliter- ation, from Amy to 艾米. Each row presents a group of corresponding subsequences in different representa- tions. bridge this gap, we define a phonetic auxiliary task that shares the sound information with the main transliteration task under the multi-task learning (MTL) setting. Depending on the specific language, the written form of a word reveals its pronunciation to various extents. For alphabetical languages such as English and French, a letter, or a sequence of letters, usually reflects the word pronunciation. For example, the word Amy (in the International Phonetic Alphabet, IPA, /"eI.mi/) has the sub-word A corresponding to /"eI./ and my corresponding to /mi/. In contrast, characters in a logographic2 writing system for lan- guages like Chinese or Japanese do not explicitly indicate sound (Xing et al., 2006). In this paper, we give a treatment to the problem of transliteration from English (alphabet) to Chi- nese3 (logogram) using an RNN-based MTL model with a phonetic auxiliary task. We transform …

Content truncated to 500 words