Translation method, apparatus and device, and storage medium

Abstract

Embodiments of the invention provide a translation method, apparatus and device, and a storage medium. The method comprises the steps of identifying entity text data of to-be-translated original textdata; performing entity translation on the entity text data; when each piece of first translation data of each piece of entity text data which is successfully translated is obtained; determining entity replacement text data corresponding to the successfully translated entity text data, and replacing the entity text data corresponding to the entity replacement text data in the to-be-translated original text data with the entity replacement text data to obtain to-be-translated replacement text data; translating the to-be-translated replacement text data to obtain second translated text data; andrespectively replacing each piece of third translation data corresponding to each piece of first translation data in the second translation data by each piece of first translation data to obtain translation text data. The accuracy of the translation result can be improved.

Description

technical field [0001] The embodiments of the present invention relate to the computer field, and in particular to a translation method, device, equipment and storage medium. Background technique [0002] Machine translation refers to a translation method that uses computers to convert between languages. [0003] In order to realize machine translation, the rule-based translation model requires professionals to summarize a large number of language habits, and then convert it into code for implementation. However, due to the complexity of the language, the rule-based translation model has low versatility and poor translation effect. [0004] For this reason, with the development of deep learning technology, a neural network-based machine translation model has appeared, including an encoder and a decoder. The encoder realizes the encoding of the translated text of the source language type, and then uses the decoder to decode the encoded feature vector. into text in the target...

AI Technical Summary

Problems solved by technology

[0005] However, despite the high versatility and translation effect of neural network-based machine translation models, the accuracy of translation results still needs to be improved

Images