Automatic Speech Recognition in CALL systems: The essential role of adaptation
暂无分享,去创建一个
Recent advances in speech technology research coupled to the increasing importance of speaking proficiency as an essential skill to acquire by migrant workers learning a second language (L2) have led to a growing interest in developing CALL systems that make use of Automatic Speech Recognition (ASR) for practicing oral proficiency. Such systems offer L2 learners the opportunity of practicing speaking outside the language classroom, at their own pace, without time limitations, and in a stress-free environment. However, direct experience with ASR-based CALL applications reveals that developing effective systems is fraught with difficulties. One of the main challenges is adapting the technology in such a way that it can cope with the idiosyncrasies of non-native learner speech while at the same time detecting the errors with sufficient accuracy.
[1] Helmer Strik,et al. Comparing different approaches for automatic pronunciation error detection , 2009, Speech Commun..
[2] Joost van Doremalen,et al. Developing a CALL system for practicing oral proficiency: how to design for speech technology, pedagogy and learners , 2009, SLaTE.
[3] Helmer Strik,et al. The goodness of pronunciation algorithm: a detailed performance study , 2009, SLaTE.
[4] Joost van Doremalen,et al. Utterance verification in language learning applications , 2009, SLaTE.