Poem recitation evaluation method, system, terminal and storage medium

Abstract

The present invention provides a poem recitation evaluation method, a poem recitation evaluation system, a terminal and a storage medium. The method includes the following steps that: a client determines a poem to be recited; the speech of the poem to be recited is obtained; the speech is preprocessed; the preprocessed speech is uploaded to a server; a pre-established recognition model in the server is adopted to convert the speech into a text; the text is compared with the poem to be recited, and the score of the text is calculated according to a preset scoring rule; and the score is fed back to the client.

Description

technical field [0001] The invention relates to the technical field of voice recognition and evaluation, in particular to a method and system for evaluating poetry reciting, a terminal and a storage medium. Background technique [0002] At present, the main users of poetry recitation are mainly primary and middle school students. It is difficult for general speech recognition technology to achieve the optimal effect of vocalization data like primary and middle school students for other users, and the recognition accuracy rate of poetry recitation is relatively low; if you want to achieve a higher High recognition accuracy requires a large amount of training data. The training data used are basically recorded by professional microphones. After a relatively long training time, the training method is also relatively complicated; now in the era of mobile Internet, users are reciting poems During detection, most of them use PC microphones, mobile phone microphones and other recor...

AI Technical Summary

Problems solved by technology

[0002] At present, the main users of poetry recitation are mainly primary and middle school students. It is difficult for general speech recognition technology to achieve the optimal effect of vocalization data like primary and middle school students for other users, and the recognition accuracy rate of poetry recitation is relatively low; if you want to achieve a higher High recognition accuracy requires a large amount of training data. The training data used are basically recorded by professional microphones. After a relatively long training time, the training method is also relatively complicated; now in the era of mobile Internet, users are reciting poems During detection, most of them use PC microphones, mobile phone microphones and other recording equipment to record poetry reciting voices, and then perform recognition and detection. Using existing voice recognition technology to recognize such poetry reciting voices recorded by non-professional microphones is not accurate. In addition, it is necessary to separately process the noise in the voice of poetry reciting, adding additional processing time for denoising, high cost and long response time

Images