Multi-language machine intelligent auxiliary processing method and system

Abstract

The invention discloses a multi-language machine intelligent auxiliary processing method and a system. The method comprises the following steps of acquiring a translated interval of a source language according to the source language and a translated target language so as to expand an untranslated interval of the source language; generating a K-best prompting set according to the expansion of the untranslated interval of the source language; generating an input historical prompting set and a language model prompting set according to the untranslated interval of the source language and the translated target language; and generating a candidate set of a target language corresponding to the untranslated source interval of the source language according to the input historical prompting set, the language model prompting set and the K-best prompting set. Due to the adoption of the method and the system, the multi-language translation speed is accelerated, the user experience is good, the intelligent degree is high, and the translation accuracy is improved.

Description

technical field [0001] The present invention relates to the technical field of natural language computer processing, in particular to the technical field of machine translation, in particular to an intelligent auxiliary processing method and system for multilingual machine translation. Background technique [0002] In the process of text translation, translators usually need to translate one language (source language) into another language (target language). [0003] In recent years, computer-aided translation based on machine translation and translation memory has solved some problems in human translation, but computer-aided translation usually has three problems. First, the granularity of word-based auxiliary translation hints is too small, and the translation of some words in sentences is often not the target language candidate directly corresponding to the word. Second, assisted translation in the form of editing after machine translation based on mobile sequencing is t...

AI Technical Summary

Problems solved by technology

First, the granularity of word-based auxiliary translation prompts is too small, and the translation of some words in sentences is often not the target language candidate directly corresponding to the word

Second, assisted translation in the form of editing after machine translation based on mobile sequencing is time-consuming and laborious, and it is very inconvenient for translators to operate

First, existing methods are usually designed directly for the target language without utilizing information from the source language; second, existing methods do not interactively combine human and machine translation; third, existing methods There are few functions that support translation, and the granularity of translation is not enough

Fourth, the existing methods cannot combine the input target language context and the translator's translation habits for intelligent translation

Images