EUROCALL 2014

Full Program »

Diagnostic CALL tool for Arabic Learners

Nowadays, learning a foreign language is an essential activity for many people. Proficiency in a foreign language is based on four different skills, namely reading skill, listening skill, writing skill, and pronunciation or speaking skill. We are particularly interested in pronunciation skill, and our proposed work is aimed at teaching non-native Arabic speakers how to improve their pronunciation. This paper reports on a CALL tool for helping non-native speakers of Arabic improve their pronunciation, particularly of words involving sounds that are not distinguished in their native languages. This tool involves the implementation of several substantial pieces of software. The first task is to ensure the system we are building can distinguish between the more challenging sounds when they are produced by a native speaker, since without that it will not be possible to classify learners’ attempts at these sounds. To this end we carried out a number of experiments with the HTK, a well known speech recognition toolkit, in order to ensure that it can distinguish between the confusable sounds, i.e. the ones that people have difficulty with.
The tool analyses the differences between the user’s pronunciation and that of a native speaker by using a grammar of minimal pairs, where each utterance is treated as coming from a family of similar words. This enables us to categorise learners’ errors—if someone is trying to say cat and the recogniser thinks they have said cad then it is likely that they are voicing the final consonant when it should be unvoiced. Extensive testing shows that the system can reliably distinguish such minimal pairs when they are produced by a native speaker, and that this approach does provide effective diagnostic information about errors.
The tool provides feedback in three different forms: as an animation of the vocal tract, as a synthesized version of the target utterance, and as a set of written instructions. We have evaluated the tool by placing it in a classroom setting and asking 30 Arabic students to use the different versions of the tool. Each student had a thirty minute session with the tool, working their way through a set of pronunciation exercises at their own pace. Preliminary results on this pilot group show that their pronunciation does improve over the course of the session, though we have not been able to determine whether the improvement is sustained over an extended period. The evaluation has been done from three points of view: quantitative analysis, qualitative analysis, and questionnaire. Firstly, the quantitative analysis gives raw numbers telling whether a learner improving his pronunciation or not. Secondly, the qualitative analysis shows a behavior pattern of what a learner did and how he used the tool. Thirdly, the questionnaire gives us a feedback from a learner and his comments about the tool.

Author(s):

Majed Alsabaan    
School of Computer Science
The University of Manchester
United Kingdom

PhD researcher in Manchester University, School of Computer Science.

Allan Ramsay    
School of Computer Science
The University of Manchester
United Kingdom

Professor in Formal group, School of Computer Science

 

Powered by OpenConf®
Copyright ©2002-2013 Zakon Group LLC