One of the most difficult aspects of the design and implementation of CAPL is its interdisciplinary nature. Particularly because most researchers tend to apply a monolithic approach to individual problems. The disciplines that have been applied to the system are as follows:
Automatic scoring does not always correlate well with human ratings. Moreover, there is no exact answer to the question ``What is the criterion of a good pronunciation?''. A reliable assessment for the quality of pronunciation had to be devised.
Most learners cannot figure out how to correct their articulation even if critical errors of their pronunciation are pointed out. More specifically, the problem how to give them feedback should be considered for the errors that result from linguistic gaps (L1 and L2) in pronunciation. Steven Hiller et al suggested the feedback method for teaching monophthong vowels within a standard context[9]. No normative methods for consonant sounds have been produced within this field.
Distinguished from the segmental aspects such as consonants and vowels, the suprasegmental aspects - rhythm, stress and intonation - are also essential to develop fluent, natural L2, especially in the production of connected speech. They are easily transferred from L1 causing L2 interference. Therefore, these suprasegmental aspects of pronunciation should be considered.
