Voice recognition method and device, computer and storage medium

Abstract

The invention discloses a voice recognition method and device, a computer and a storage medium. The method comprises the steps that testing voice data input in the product defect testing process is acquired; the testing voice data is subjected to voice recognition through a voice recognition model, and the recognition text of the testing voice data is acquired; a trained Markov model is obtained;the recognition text is input into the Markov model, the recognition text is recognized through the Markov model, and the output probability of all correction texts output by the Markov model are acquired; the correction text corresponding to the recognition text is determined according to the output probability of the correction texts. According to the proposal, the Markov model is introduced, the recognition text obtained after common voice recognition is carried out on the testing voice data is corrected, then the correction text corresponding to the testing voice data is obtained, the efficiency of a tester of submitting testing defect data is improved, and the user experience is improved.

Description

technical field [0001] The invention relates to the field of speech processing, in particular to a speech recognition method, device, computer equipment and storage medium. Background technique [0002] Nowadays, when submitting software test defect data to the server, it is necessary to manually upload information such as text, pictures, or logs. Vocabulary cannot be recognized by ordinary speech recognition models, which will result in that after the audio is converted into recognized text, the recognized text does not match the test defect data. Therefore, a speech recognition method is needed that can accurately recognize specialized vocabulary in the test field The test enables voice data, so that users can submit software test defect data by voice, which improves the efficiency of submitting software test defect data and enhances user experience. Contents of the invention [0003] Based on this, the present invention provides a voice recognition method, device, comp...

AI Technical Summary

Problems solved by technology

[0002] Nowadays, when submitting software test defect data to the server, it is necessary to manually upload information such as text, pictures, or logs. Vocabulary cannot be recognized by ordinary speech recognition models, which will result in that after the audio is converted into recognized text, the recognized text does not match the test defect data. Therefore, a speech recognition method is needed that can accurately recognize specialized vocabulary in the test field The voice data of the test enables the user to submit the defect data of the software test by voice, which improves the efficiency of submitting the defect data of the software test and enhances the user experience

Images