Text-to-Speech

Definition
Text-to-Speech is a system of speech synthesis which transforms written text into speech. A number of components can be distinguished: such as the linguistic component, the prosodic component, and the phonetic component. Several analyses are involved in the linguistic component: e.g. syntactic analysis of sentences, and morphological decomposition (in order to split the word up in morphemes and to predict lexical stress). Then there is the phonetic component which takes care of the prosody (i.e. sentence accent, temporal organisation, intonation contour), the grapheme-phoneme conversion, and assimilation and coarticulation rules.

Links

 * Utrecht Lexicon of Linguistics
 * Introduction to Text-to-Speech (Click 'Short introduction to TTS synthesis')