将正字法音节作为相关语言间统计机器翻译的基本单元
计算与语言
2016-10-04 v1
摘要
我们探索将正字法音节（一种变长的辅音-元音序列）作为使用元音附标文字或字母文字的相关语言之间翻译的基本单元。我们表明，在小型平行语料库上进行训练时，正字法音节级翻译显著优于在其他基本单元（词、语素和字符）上训练的模型。
引用
@article{arxiv.1610.00634,
title = {Orthographic Syllable as basic unit for SMT between Related Languages},
author = {Anoop Kunchukuttan and Pushpak Bhattacharyya},
journal= {arXiv preprint arXiv:1610.00634},
year = {2016}
}
备注
7 pages, 1 figure, compiled with XeTex, to be published at the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2016