SBIR Phase I: A Software Tool for Teaching Reading Based on Text-to-Speech Letter-to-Phoneme Rules

This Small Business Innovation Research Phase I project proposes to develop software for interactive teaching of reading, based on letter-to-phoneme rules developed for an entirely different technology, namely computer text-to-speech synthesis (TTS). Awareness of letter-phoneme correspondences is important for learning to read English. However, no tools are available that identify and apply all the letter-to-phoneme rules occurring in any word or name of English. Challenges are to identify which letter-to- phoneme rules are useful for the learner, to determine which words are best presented as exceptions, how to present letter-to-phoneme rules and exceptions effectively for learners, and how to integrate the software into an overall reading program. Phase I will develop a software prototype, criteria for identifying and formulating useful rules, and quantitative methods for assessing the effectiveness of the software. The role of explicit reference to the letter-sound correspondences in the teaching of reading has experienced fluctuations in popularity, in part due to the lack of educators' unanimity about which of the rules that can be deduced are real and helpful. E-Speech's letter-to-phoneme rules have undergone decades of development and serve as an excellent basis for identifying useful word pronunciation rules for learners. The resultant software would enable beginning readers, adult learners, learners of English as a second language, and readers with learning disabilities to learn English word pronunciation, and it could be used by teachers who find students having difficulty with certain words. The technology would provide reading assistance to users on an individual basis, it could allow users to learn in a private setting, and it could be cost-effective, since it could reduce the amount of time required with a human teacher. The technology could be incorporated into other educational software packages and programs. Consequently, it would contribute substantially to gains in American literacy as well as support the internationalization of English. In addition, the software might also become part of a computer-based, standard dictionary of English, supplementing the phonetic transcriptions for all words in a dictionary.