Many students of English language find pronunciation difficult to master. This work in progress paper discusses an incremental and iterative approach towards developing requirements for software applications to assist learners with the perception and production of English pronunciation in terms of phonemes and prosody. It was found that prompts for eliciting target pronunciation should include a visual indication of the meaning of the target word or phrase in addition to the sound, and that the learners should be led through a hierarchy of words. This should start with phonemes of simple (one syllable) words, and adaptively build up to prosody of two syllable words, then increasing the syllables in the target words as the learner improves. A simple representation of prosody was developed and found to be intuitive by students for comparing their pronunciation to that of a native speaker. Students considered that an analysis time of within one second for phonemes and prosody was considered "real time", and requested integration with social media for both enabling competition and celebration of achievement. [For full proceedings, see ED565087.]