Neural machine translation model training methods, devices and storage media

摘要

PROBLEM TO BE SOLVED: To provide a training method, an apparatus and a storage medium for a neural machine translation model. The method of training a neural machine translation model provided by an embodiment of the present invention trains N grams, which are commonly found in a target sentence corpus, as indivisible words in the process of training a neural machine translation model. This results in a trained neural machine translation model containing more N grams than the actual translation. As a result, the scoring result of the neural machine translation is improved, and the machine translation quality is improved. [Selection diagram] Fig. 1