Text To Speech Program

Simple phoneme synthesis program
Page last updated on 5/31/2021

In high school, I took an elective course called "Computer Science III" where I was allowed to independently research topics of my choosing and write programs related to them. I decided to write a text to speech program, and use it as an opportunity to strengthen my knowledge of C++ file I/O and pointers (which was limited at the time, I've learned a lot since then!) My program works by requiring the user to make audio recordings of them uttering all the unique phonemes of the English language (there's about 54 of them, when including different emphases), and using these audio recordings to synthesize words and sentences. The user can write a sentence on the command line, and each word is converted to a set of phonemes using a word-to-phoneme dictionary I found on the Internet, and the resultant string of phonemes are then sequenced into a single .WAV file which is outputted. I recorded myself uttering the ~54 phonemes and found the results to be very robotic, but still surprisingly intelligible!