Coursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures Translation

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 3640–3649 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 3640 Coursera Corpus Mining and Multistage Fine-Tuning for Improving Lectures Translation Haiyue Song1, Raj Dabre2, Atsushi Fujita3, Sadao Kurohashi4 1,4Kyoto University 2,3National Institute of Information and Communications Technology 1song@nlp.ist.i.kyoto-u.ac.jp, 4kuro@i.kyoto-u.ac.jp 2,3{raj.dabre, atsushi.fujita}@nict.go.jp Abstract Lectures translation is a case of spoken language translation and there is a lack of publicly available parallel corpora for this purpose. To address this, we examine a framework for parallel corpus mining which is a quick and effective way to mine a parallel corpus from publicly available lectures at Coursera. Our approach determines sentence alignments, relying on machine translation and cosine similarity over continuous-space sentence representations. We also show how to use the resulting corpora in a multistage fine-tuning based domain adaptation for high-quality lectures translation. For Japanese–English lectures translation, we extracted parallel data of approximately 40,000 lines and created development and test sets through manual filtering for benchmarking translation performance. We demonstrate that the mined corpus greatly enhances the quality of translation when used in conjunction with out-of-domain parallel corpora via multistage training. This paper also suggests some guidelines to gather and clean corpora, mine parallel sentences, address noise in the mined data, and create high-quality evaluation splits. For the sake of reproducibility, we have released our code for parallel data creation. Keywords: parallel corpus, machine translation, educational corpus, alignment, lecture translation, fine-tuning 1. Introduction In recent years, massive open online courses (MOOCs) have proliferated and have enabled people to attend lectures regardless of their geographical location. Typically, such lectures are taught by professors from certain universities and are made available through video recordings. It is com- mon for these lectures to be taught in a particular language and the video is accompanied by subtitles. These subtitles are then translated into other languages so that speakers of those languages also benefit from the lectures. As manual translation is a time-consuming task and there are a large number of lectures, having a high-quality machine transla- tion (MT) system can help ease the distribution of knowl- edge to a large number of people across the world. Given the fact that most lectures are spoken in English, translating English subtitles to other languages is an urgent and important task. On the other hand, there are also lec- tures taught in other languages than English. For instance, several universities in Japan offer online lecture courses, mostly taught in Japanese. Enabling non-Japanese speak- ers to participate in these courses by translating Japanese lecture subtitles into other languages, including English, is also an important challenge. The TraMOOC project (Ko- rdoni et al., 2015) aims at improving the accessibility of European languages through MT. They focus on collecting translations of lecture subtitles and constructing MT sys- tems for eleven European and BRIC languages. However, the amount of parallel resources involving other languages, such as Chinese and Japanese, are still quite low. Subtitle translation falls under …

Content truncated to 500 words