Is Neural Machine Translation Ready for Deployment? A Case Study on 30 Translation Directions

10/04/2016 ∙ by Marcin Junczys-Dowmunt, et al. ∙ 0

In this paper we provide the largest published comparison of translation quality for phrase-based SMT and neural machine translation across 30 translation directions. For ten directions we also include hierarchical phrase-based MT. Experiments are performed for the recently published United Nations Parallel Corpus v1.0 and its large six-way sentence-aligned subcorpus. In the second part of the paper we investigate aspects of translation speed, introducing AmuNMT, our efficient neural machine translation decoder. We demonstrate that current neural machine translation could already be used for in-production systems when comparing words-per-second ratios.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 3

page 4

Code Repositories

marian

Fast Neural Machine Translation in C++


view repo

nazgul

The Translation Engine Wrapper and Servant


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.