Automatic Assessment of Non-Native Prosody – Annotation , Modelling and Evaluation
Automatic Assessment of Non-Native Prosody – Annotation , Modelling and Evaluation
复制标题
非母语韵律的自动评估——注释、建模和评估
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Florian Hönig June
中科院分区：
文献类型：
--
作者：
Florian Hönig June
第一部分涉及对用于发音评估的人工评分器和自动系统的评估的一般考虑：考虑到人工注释器的易错性和机器学习算法(和研究人员)的性质，我们如何才能更接近对其可靠性的公正、现实的估计，这些算法适应并不可避免地过度适应给定的训练集。在第二部分，我们将提供具体的模型来评估学习者演讲的整体节奏质量。在德国研究项目C-AUDIT和AUWL的阅读和半自发的英语数据上，对这些方法进行了详细的评估。
The first part deals with general considerations on the evaluation of both human raters and automatic systems, employed for pronunciation assessment: How can we come closer to an unbiased, realistic estimate of their reliability, given the fallibility of human annotators and the nature of machinelearning algorithms (and researchers) that adapt, and inevitably overfit to a given training set. In the second part, we will present concrete models for the assessment of the overall rhythmic quality of the learner’s speech. The methods are evaluated in detail on read and semi-spontaneous English data from the German research projects C-AuDiT and AUWL.