Post-machine-translation term translation method and device, equipment and storage medium

Abstract

The invention provides a post-machine-translation term translation method. The post-machine-translation term translation method comprises the following steps that: a first term library and a second term library are created for an original text; the terms with the one-to-one original translation relation in the original text are stored in a first term library, and the terms with the one-to-many original translation relation in the original text are stored in a second term library; machine translation is performed on the original text to generate a translated text and original text and translated text alignment information; on the basis of the original text translation alignment information, the first term library is used to replace one-to-one term translations in the translation text with term translations in the first term library; and based on the original text and translation alignment information and a first operation instruction of the user, a second term library is used to at least indicate one-to-many term original texts in the original text, and indication information is generated. The invention further provides a post-machine-translation term translation device, electronic equipment and a storage medium.

Description

technical field [0001] The disclosure relates to a term translation method, device, equipment and storage medium after machine translation, and belongs to the technical field of translation. Background technique [0002] Although the quality of neural network machine translation has been greatly improved compared with rule-based and statistical machine translation, when translating content with high precision requirements, especially in professional fields, human translation is still required to intervene and correct. [0003] Post-translation editing of machine translation has gradually become an important translation method. On the one hand, it greatly improves translation efficiency, and on the other hand, it ensures translation accuracy. However, post-editing content generally has higher requirements and requires precise translation. In particular, how to accurately and efficiently translate the terms in the translated content has always been an important issue in the fi...

AI Technical Summary

Problems solved by technology

[0005] Term translation based on machine translation, which can achieve term replacement in purely automatic translation process, but cannot add terminology again in assisted translation or post-editing process

[0006] Based on the terminology prompt of the auxiliary translation software, when the translator translates the relevant content, the relevant term in the term base can be automatically queried and prompted, but it cannot be replaced automatically. The translator needs to manually select and perform the replacement operation

In the case of a large amount of translation content and a large amount of terms, there is a problem of low efficiency

[0007] In the term base, there is often a situation where one original term corresponds to multiple translations of terms. When replacing a term, it may not be possible to determine which translation of the term should be replaced, because only one term can be reserved, and there may be a substitution error, resulting in a translation Inaccurate

In the case of multi-person collaboration, the terms added by the collaboration members cannot refer to each other, resulting in the loss of the translated version of the terms contributed by the collaboration members and waste of resources

Images