Method and device for recognizing personalized speeches

Abstract

The invention provides a method and a device for recognizing personalized speeches. The method includes: A, determining whether a speech to be recognized belongs to an authorized user or not; if yes, using a speech recognition module corresponding to the authorized user to recognize the speech to be recognized; and if not, executing a step B; B, determining what dialect type the speech to be recognized belongs to, and using a speech recognition module, corresponding to the dialect type which the speech to be recognized belongs to, to recognize the speech to be recognized. Precision in recognizing speeches of various users can be improved by the use of the method and device.

Description

【Technical field】

[0001] The invention relates to speech recognition technology, in particular to a personalized speech recognition method and device. 【Background technique】

[0002] Speech recognition technology has important applications in many fields, such as voice dialing, voice navigation and other applications, all rely on speech recognition technology. The effect of speech recognition directly affects the effect of various applications related to speech recognition.

[0003] The ability of the speech recognition model to describe speech to text is a key factor that affects the quality of speech recognition. The existing technology uses the same speech recognition model to recognize the speech of various users. Due to the same speech recognition model, it is difficult to describe the differences in the speech of different users. Therefore, using this method for speech recognition, the recognition accuracy in many cases Poor, it is difficult to meet the individual ne...

Images