Grapheme-to-phoneme

I do not really prefer English text. The badly distorted grapheme-to-phoneme relationships are really ugly. Silent letters, double letters, distorted diphthongs, awkward letter sequences, mumbled syllables, etc. are really disorganization. Because, being a linguist, I can compare with other languages, I am aware of these things.

Automatic grapheme-to-phoneme conversion was first considered in the context of text-to-speech (TTS) applications. After normalization (expanding abbreviation, numerals, etc.) the input text needs to be converted to a sequence of phonemes which is then used to control a speech synthesizer. The simplest technique is dictionary look-up.