基于隐马尔可夫模型的英语口语考试智能评分系统

Intelligent Scoring System of Spoken English Test Based on Hidden Markov Model

针对英语口语考试评分系统工作过程复杂、评分准确度低的问题,构建基于隐马尔可夫模型的英语口语考试评分系统.设计了语音识别模块和评分流程,采用语音识别器和隐马尔可夫评分模型,实现对英语口语发音的识别、错误判定和评分,并由百分制变换得到英语口语的最终得分.实验结果表明,该系统的评估准确度高,可提高系统错误判定的准确率和稳定性.

In view of the complexity of scoring system and the low accuracy of scoring system in spoken English test,a scoring system for spoken English test is built based on Hidden Markov model. The speech recognition module and scoring process are designed, a speech recognizer and a hidden Markov scoring model are adopted, to achieve pronunciation, recognition, and grading of spoken English and the final score of spoken English is changed by 100 percent system. Experimental results show that the pro- posed system has high accuracy and can improve the accuracy and stability of the system error detection.