Remote voice recognition system and method with degree of association

Abstract

The invention discloses a remote voice recognition system and method with the degree of association, and belongs to the technical field of voice recognition. The system includes an intelligent device and a server side. The intelligent device includes an input unit, a control unit, a storage unit and a registration unit. The server side includes an updating unit, a receiving unit, a judgment unit and a recognition unit. The intelligent device uploads recognition models to the server side to complete registration, and uploads a voice signal input by a user to the server side; and the server side updates a standard recognition model according to the successfully registered recognition models, and then according to the standard recognition model, a recognition instruction corresponding to an expected voice value that is equal to the voice signal serves as a recognition result to be returned to the intelligent device; and the intelligent device processes the association between the expected voice value and the recognition instruction according to the recognition result and executes corresponding operation. Voice recognition management of different intelligent devices planned as a whole can be realized, and the accuracy rate of voice recognition is improved, thereby improving the usage experience of the user.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to a remote speech recognition system and method provided with a degree of relevance. Background technique [0002] With the development of science and technology, more and more smart devices will integrate voice control functions to facilitate users to control the smart devices, such as integrating voice control functions on smart robots. [0003] In the prior art, for different smart devices, since the target users may be different, the recognition models that need to be used therein will also be different, so when performing speech recognition in different smart devices, it is very Recognition errors may be caused by different recognition models, and the correlation between the expected voice value and the recognition instruction in the recognition model is also very important. That is, there is no unified voice recognition management solution in the prior art to contr...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, for different smart devices, since the target users may be different, the recognition models that need to be used therein will also be different, so when performing speech recognition in different smart devices, it is very Recognition errors may be caused by different recognition models, and the correlation between the expe...

Images