Make Neural Machine Translation Better, Faster
ONEs - OHT NMT Evaluation Score, is a new way to Measure NMT Quality.
ONEs is calculated for 10 types of material – general, legal, marketing, finance, gaming, software, medical, technical, scientific, and tourism texts
Compare 10 NMT engines
Which NMT engine is best for your content?
Google, Microsoft (Bing), Amazon, DeepL, Systran, Baidu, Promt, IBM Watson, Globalese and Yandex
Statistically Significant, Always Up-to-date
At least 40 bilinguals rate every translation. The score is recalculated every quarter.
We have recruited, trained, and tested thousands of linguists in over 50 languages, and already run well over 1,000,000 NMT rating and testing projects for our customers.
The scores are used in real-time to produce high-quality "hybrid" translation, combining NMT and human post-editing.
Better than any algorithm
While BLEU, METEOR, and TER are fast, easy, and inexpensive to run, they are just not good enough to evaluate NMT. Why not using a real human being?