Research On Oral English Evaluation Algorithms

The evaluation of pronunciation for spoken English is one of the key problems for computer aided spoken language learning. Though current speech pronunciation evaluation system provides a lot of encouraged results, they usually pay more attention on phonetic features of the pronunciation, and few of them pay attention to the using of specific English grammars. Though waveform comparison or other kinds of feedbacks have been provided to language learners, few of them provide specific and very useful evaluation results for the learners to improve given aspects of their spoken language skills. This paper mainly deals with the evaluation of liaison and HDP (phoneme that is hard to be distinguished) in spoken English.First, it is the evaluation model of liaison in spoken English. Liaison evaluation for spoken English is one of the key problems for computer aided spoken language learning. Though a lot of factors affect the performance of a spoken language evaluation algorithm, there are mainly two factors that contribute to most of the obstacles, i.e. the natural casualness of spoken language and the unstable performance of existed speech processing systems. In this paper, the Sugeno integral is introduced to address the problems faced in liaison evaluation. It models the casualness of the spontaneous speech sound and the unstableness of the speech processing system via fuzzy measures and reliabilities respectively, and then fuses them under the Sugeno integral framework to output the linguistic rather than quantitive score. The experiment shows that, under the 58% average recognition rate of the system, our liaison evaluation model get very reliable and stable results for both closing and opening test, which encourage us to apply this model to other tasks of the spoken language evaluation in the future.Second, it is the evaluation model of HDP. While most of researchers focuse on the improvement of speech recognition to build a reliable evaluation system, there still needs a model that fuses the reliabilities of existed speech processing systems and the learner's personalities into the evaluation system. In this paper,...