A text-to-speech environment: applications in education

The authors discuss the design and implementation of a text-to-speech (TTS) system on the NeXT computer. The TTS system accepts as input any unrestricted text file which is then normalized. The phoneme generation module parses this normalized text in order to isolate word chunks, which are used to produce the corresponding phonetic transcription. The phonemes generated for a particular word are stored in an array, which is then sent to the allophone generation module. The refinement of pronunciation takes place in the allophone generation module which also sends the allophones to the digital signal processor. Applications of the TTS system to education are discussed