The ESCA Workshop on Speech Synthesis September 25-28, 1990 Autrans, France

From Orthography to Phonetic Transcription in the German Text-To-Speech System Tetos

Klaus Wothke

IBM Scientific Center, Heidelberg, Germany

A text-to-speech system was developed which reads aloud unrestricted German texts. It converts orthographic input text into synthetic speech in 3 main steps: First the words of the text are preprocessed and abbreviations, special characters, and digits are replaced by their full orthographic correlates. Then the preprocessed words are phonetically transcribed by means of about 1,370 letter-to-phone rules. Finally the phonetic transcriptions are mapped on sequences of control codes for a speech synthesizer. The rate of incorrect conversions of written words to phonetic transcriptions is near 2% of the running words of a text.

Full Paper

Bibliographic reference.  Wothke, Klaus (1990): "From orthography to phonetic transcription in the German text-to-speech system tetos", In SSW1-1990, 219-224.