Bertjan Busser

Project: Machine Learning algorithms for linguistic aspects of speech synthesis

Basically I am to develop a (properly researched) Grapheme-to-Phoneme Convertor capable of generating suprasegmental prosodic information. Grapheme-to-Phoneme Conversion is translating (ASCII) text into its phonetic transcription (the pronunciation cues in the dictionary. Suprasegmental prosodic information is usually paraphrased speech melody, and involves rhythm, intonation, and pauses.


I studied Electrotechnics at Twente University (1988-1990) and at the Hogeschool Enschede (1990-1992). The next two years ('till the end of '94) I studied (theoretical) Psychology at the Utecht University and graduated on a thesis on deictic reference in spatial localization. I worked 9 months in a revalidation research centre on categorizing signs from Dutch Sign Language before I started on this project.

Last modified: Mon Aug 2 13:16:39 CEST 1999