Multi-page engine collaborative speech translation system and method based on reverse text consensus

Abstract

The invention provides a multi-page engine collaborative speech translation system and method based on reverse text consensus. The system comprises: a voice input subsystem used for inputting voice to be translated; the first speech translation subsystem that is used for translating the speech to be translated and outputting a first speech translation result; a second text translation subsystem; a consensus judgment subsystem; and an output subsystem. The second text translation subsystem comprises a plurality of text translation engines; each first speech translation result output by the first speech translation subsystem serves as input of a plurality of text translation engines of the second text translation subsystem, and the second text translation subsystem outputs a plurality of text translation results; the consensus judgment subsystem performs translation result consensus judgment based on the plurality of text translation results; and the output subsystem outputs a translation record file corresponding to the voice to be translated based on translation result consensus judgment. The invention further provides a speech translation method based on the system.

Description

technical field [0001] The invention belongs to the technical field of speech intelligent translation, and in particular relates to a multi-translation engine cooperative speech translation system and method based on reverse text consensus. Background technique [0002] With the increasing globalization of human society, economy and culture, people with different native languages ​​communicate more and more frequently in many occasions such as tourism, conferences, business, sports, etc., and language barriers are becoming more and more obvious. Therefore, people urgently hope that computers can have higher intelligence to act as an intermediate role to overcome human language barriers and realize free communication between people. This kind of translation from one natural language voice to another natural language voice realized by computer system is what we usually call Speech-to-Speech Translation. Since the daily communication between people is usually carried out in th...

AI Technical Summary

Problems solved by technology

However, because meetings usually need to maintain real-time performance and avoid long delays, the existing systems can only directly save the speech records of all participants, which obviously increases the amount of data storage and transmission, especially in remote meetings. Larger transmission costs

If all speeches are not recorded, it is necessary to manually proofread the translation results in different languages, which reduces work efficiency and increases labor costs

Images