Hi! Have you ever considered that one set of English sounds can be produced by
more than 25 variations in letters depending on the origin of the word, adjoining letters, and grammatical context?
A vivid example:
[gi:] can be produced by
GGY,GEY,GHEE,GGEE,GAE,GEA,GEE,GE,GHI,GIE,GI,GUI,GYI,GGIE,GGI,GY,GHEY,GGE,GHE,GII.Ah, a child’s brain does great work to acquire language skills.
Well, school days have arrived, and it’s high time to help a kid read. I’d like to ask advice from experts, regarding an
untraditional method of learning to read. What do you think, does dividing words into
easy-to-pronounce parts (such as consonant+vowel, one vowel, or one consonant) make the learning process easier? For example, the word “ELMO” consists of 3 parts:
E-L-MO, DO-NU-T, CA-T, SEA-SI-DE.
To answer the question, I wrapped this approach into a mobile app. It’s like a phonics keyboard which can voice any typed word, and then divide it into the parts.
To understand this better, you can try it here:
Appstore PlayMarketI’m an indie developer and
greatly appreciate your experience! I look forward to your advice.
BTW, what method do you prefer?
Thanks!