Translation method, device and equipment and storage medium

Abstract

The invention discloses a translation method and device, equipment and a storage medium, and the method comprises the steps: obtaining a source language text, voice information of the source language text and structural information of each text subunit in the source language text at the same time, and respectively carrying out the feature extraction of the source language text, the voice information and the structural information; and obtaining text features corresponding to the source language text, voice features corresponding to the voice information and structural features corresponding to the source language text, and then performing translating based on the three features to obtain a translated target language text. Therefore, in the translation process, the text, the pronunciation and the structure serve as different cognitive levels of the text of the same source language to be comprehensively considered, translation is conducted based on the three characteristics, full utilization of resource information is achieved, and meanwhile, due to the fact that information referred during translation is richer, the robustness of the translation process can be improved, and the quality of translation results can be improved.

Description

technical field [0001] The present application relates to the technical field of machine translation, and more specifically, relates to a translation method, device, equipment and storage medium. Background technique [0002] In recent years, the staged breakthroughs in technologies such as deep learning and transfer learning, as well as the large amount of data generated by the Internet, have brought a lot of opportunities for the development of natural language processing, and breakthroughs have also been made in natural language processing tasks. Among them, machine translation is a natural One of the more important applications in language processing tasks. With the development of natural language processing tasks, the accuracy of machine translation has also been greatly improved. The increasing reliability of using machines to automatically process text has allowed more work to be done by machines. Deep learning usually relies on a large amount of labeled data. In rec...

AI Technical Summary

Problems solved by technology

When the number of parallel sentence pairs in the translation task is relatively rich, the quality of the translation model is high, but when the sentence pair data of the translation task is limited, it also greatly limits the quality of the machine translation model

[0004] Although the availability of machine translation effects in resource-rich scenarios is already very high, there is still a lot of room for improvement in the effects of low-resource scenarios

Images