Abstract
In oral English learning, HDPs (phonemes that are hard to be distinguished) are areas where Chinese students frequently make mistakes in pronunciation. This paper studies a speech phoneme evaluation method for HDPs, hoping to improve the ability of individualized evaluation on HDPs and help provide a personalized learning platform for English learners. First of all, this paper briefly introduces relevant phonetic recognition technologies and pronunciation evaluation algorithms and also describes the phonetic retrieving, phonetic decoding and phonetic knowledge base in the Sphinx-4 computer system, which constitute the technological foundation for phoneme evaluation. Then it proposes an HDP evaluation model, which integrates the reliability of the speech processing system and the individualization of spoken English learners into the evaluation system. After collecting HDPs of spoken English learners and sorting them into different sets, it uses the evaluation system to recognize these HDP sets and at last analyzes the experimental results of HDP evaluation, which proves the effectiveness of the HDP evaluation model.
Highlights
With the implementation of national economic and political development strategies like “Going-out” and “Belt and Road”, Chinese are facing more opportunities of international communication and trade, and English, as an international language, is being more widely learnt in China [1]
Spoken English is the most direct form of English learning and communication; for most Chinese English learners, pronunciation has become their bottleneck in improving their spoken English, especially HDP, which are the areas where Chinese spoken English learners frequently make mistakes [2]
It introduces the typical computer-based automatic speech recognition technology and spoken language evaluation methods and describes the front-end, decoding and knowledge base functions of the key computer system Sphinx-4 applied in this paper [4], which has laid a theoretical and technical foundation for the spoken English phoneme evaluation algorithm
Summary
With the implementation of national economic and political development strategies like “Going-out” and “Belt and Road”, Chinese are facing more opportunities of international communication and trade, and English, as an international language, is being more widely learnt in China [1]. HDP is the symbolic feature of spoken English, reflecting the liaison of pronunciations and phonemic feature in spoken English. These two are the important guarantee for barrier-free communication with foreigners [3]. This paper aims to study the phoneme evaluation algorithm for spoken English to achieve phonetic feature evaluation on HDP. The establishment of an HDP model is of great guiding significance to the evaluation on other features of spoken English. It is an extension of computer-assisted teaching. Studying and promoting this model will help spoken English learners in China improve their English skills in individualized ways through computer technology
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
