Translating Classical Chinese Poetry into Modern Chinese with Transformer

Peng Jin, Hailiang Wang, Limin Ma, Bing Wang, Shushan Zhu

Abstract: Classical Chinese poetry, as the cultural heritage of human beings, is very popular in Chinese community all over the world. Nearly every person in these regions can recite several poems to artistically express his or her emotion. However, due to the huge difference between the archaic and modern Chinese language, now, it is hard for people to understand these poems. Considering the Neural machine translation (NMT) has made great progress since the self-attention technology called Transformer was adopted, we followed the paradigm of NMT to translate the classical Chinese poems into modern Chinese texts in this study. First, we collect many ancient poems and their explanations as a parallel corpus. Then, a NMT system is trained with this corpus. Finally, we compared this system with Generative Pre-Training (GPT) and analysis their results in detail.

External IDs:dblp:conf/clsw/JinWMWZ21