Achieving Human Parity on Automatic Chinese to English News Translation

TLDR

Machine translation has rapidly advanced, with millions of users worldwide, raising the question of whether it can achieve parity with human translations. The study aims to define and accurately measure human parity in translation. The authors describe Microsoft’s neural machine translation system and evaluate its Chinese‑to‑English output on the WMT 2017 news task. The system reaches state‑of‑the‑art performance, achieving human‑parity quality compared to professional translators and surpassing crowd‑sourced non‑professional translations.

Abstract

Machine translation has made rapid advances in recent years. Millions of people are using it today in online translation systems and mobile applications in order to communicate across language barriers. The question naturally arises whether such systems can approach or achieve parity with human translations. In this paper, we first address the problem of how to define and accurately measure human parity in translation. We then describe Microsoft's machine translation system and measure the quality of its translations on the widely used WMT 2017 news translation task from Chinese to English. We find that our latest neural machine translation system has reached a new state-of-the-art, and that the translation quality is at human parity when compared to professional human translations. We also find that it significantly exceeds the quality of crowd-sourced non-professional translations.