Modular Design of English Pronunciation Proficiency Evaluation System Based on Speech Recognition Technology
Ding Li · 2022 IEEE 2nd International Conference on Mobile Networks and Wireless Communications (ICMNWC) · 2022
With the vigorous development of computer technology, speech recognition technology (SRT) is gradually applied to various fields. Especially in foreign language teaching, SRT assists speech learning software, which can provide learners with learning resources such as multiple sounds, images, and videos. It can also help learner’s correct pronunciation and improve the correct level of English pronunciation. Therefore, SRT has strong practicability in the English pronunciation proficiency evaluation system. The research theme of this paper is the modular design of English pronunciation proficiency evaluation system based on SRT. This article first elaborates the related technologies of the English pronunciation proficiency evaluation system, and carries out modular design of the system from four aspects: intonation, speaking speed, intonation, and rhythm. Finally, quantitative audio-visual dual-modal speech recognition and cross-modal speech recognition tests were carried out on the system. According to the experiment, in the English pronunciation proficiency evaluation system the accuracy of intonation, speech speed, rhythm and intonation is as high as 88.39%, the adjacent agreement is as high as 99.76%. This means that the system is in the evaluation of English pronunciation level is feasible and effective.