A spoken pronunciation evaluation method, device, electronic equipment and storage medium

Abstract

The present invention provides a spoken pronunciation evaluation method, device, electronic equipment and storage medium, obtain the audio frame sequence corresponding to the speech segment to be evaluated and the phoneme sequence corresponding to the speech segment to be evaluated; determine the basis of the pronunciation state of each phoneme corresponding to each audio frame Label; adjust the basic label of the phoneme pronunciation state corresponding to each frame of audio frame, and obtain the final label of the phoneme pronunciation state corresponding to each frame of audio frame; according to the final label of the phoneme pronunciation state, determine the target word in the speech segment to be evaluated. The energy ratio in the speech segment; according to the energy ratio, the oral pronunciation evaluation result is obtained. The energy ratio of the target word in the speech segment to be evaluated can reflect whether there is random reading, so that according to the energy ratio, the random reading phenomenon can be detected, and the oral pronunciation evaluation results obtained according to the energy ratio can reflect the random reading phenomenon. Pronunciation evaluation results can improve the accuracy of oral pronunciation evaluation results.

Description

technical field [0001] The present application relates to the technical field of speech recognition, and more specifically, relates to a spoken pronunciation evaluation method, device, electronic equipment and storage medium. Background technique [0002] With the current economic development and cultural exchanges, the requirements for individuals' foreign language ability are getting higher and higher. In the process of learning a foreign language, how to improve the oral ability of a foreign language is an urgent problem to be solved at present. With the development of artificial intelligence, speech recognition technology has been applied to oral pronunciation practice. At present, there are many oral pronunciation evaluation software for evaluating oral pronunciation. practice. [0003] However, there are currently some users who read randomly in the oral pronunciation evaluation software. These random readings seriously affect the accuracy of the oral pronunciation e...

AI Technical Summary

Problems solved by technology

[0003] However, there are currently some users who read randomly in the oral pronunciation evaluation software. These random readings seriously affect the accuracy of the oral pronunciation evaluation results, and the current oral pronunciation evaluation methods cannot accurately detect these random readings. Based on Therefore, how to effectively detect the phenomenon of random reading and improve the accuracy of the spoken pronunciation evaluation results is a technical problem that needs to be solved at present.

Images