Voice recognition method and device

Abstract

The present invention discloses a voice recognition method and a device. The method comprises the steps of firstly, acquiring the voice information input by a speaker and acquiring the information of the speaker; secondly, judging the existence / absence of a personal acoustic model corresponding to the speaker according to the information of the speaker; thirdly, upon judging the existence of the personal acoustic model, acquiring the personal acoustic model and conducting the voice recognition on the voice information according to the personal acoustic model of the speaker; fourthly, upon judging the absence of the personal acoustic model, conducting the voice recognition on the voice information according to a basic acoustic model, generating the corpus information of the speaker according to the voice information, and storing the corpus information; fifthly, generating a personal acoustic model for the speaker based on the basic acoustic model and the stored corpus information. Based on the method, acoustic models can be customized for all speakers based on the characteristics of the speakers during the self-adaptive voice recognition process of the speakers. Therefore, the recognition accuracy for each speaker is improved and the user experience is improved.

Description

technical field [0001] The invention relates to the technical field of voice recognition, in particular to a voice recognition method and device. Background technique [0002] In recent years, speech recognition technology has developed rapidly, especially after the deep neural network is applied to speech recognition, the performance of speech recognition has been greatly improved. With the development of the mobile Internet, voice input methods are becoming more and more common, and voice users are becoming more and more extensive. Therefore, how to improve the accuracy of speech recognition has become an urgent problem to be solved. [0003] In related technologies, the speech recognition process mainly uses a large amount of speech training to obtain an acoustic model and a language model, and then performs speech recognition on speech data input by a speaker through the acoustic model and the language model. It can be seen that the larger the training sample, the high...

AI Technical Summary

Problems solved by technology

[0004] However, the existing problem is that in the above speech recognition process, a large number of speech samples are used to train and build an acoustic model. This model is applied to the speech recognition process of all speakers. For speech with a heavy dialect accent or unclear speech For humans, the above voice recognition methods may not be able to recognize the content input by the speaker very well, which reduces the recognition accuracy of the acoustic model and deteriorates the user experience.

Images